This dataset contains 10,000 simulated sales transaction records, each represented in natural language with diverse sentence structures. It is designed to mimic how different users might describe the same type of transaction in varying ways, making it ideal for Natural Language Processing (NLP) tasks, text-based data extraction, and accounting automation projects.

Each record in the dataset includes the following fields:

Sale Date: The date on which the transaction took place. Customer Name: A randomly generated customer name. Product: The type of product purchased. Quantity: The quantity of the product purchased. Unit Price: The price per unit of the product. Total Amount: The total price for the purchased products. Tax Rate: The percentage of tax applied to the transaction. Payment Method: The method by which the payment was made (e.g., Credit Card, Debit Card, UPI, etc.). Sentence: A natural language description of the sales transaction. The sentence structure is varied to simulate different ways people describe the same type of sales event.

Use Cases: NLP Training: This dataset is suitable for training models to extract structured information (e.g., date, customer, amount) from natural language descriptions of sales transactions. Accounting Automation: The dataset can be used to build or test systems that automate posting of sales transactions based on unstructured text input. Text Data Preprocessing: It provides a good resource for developing methods to preprocess and standardize varying formats of text descriptions. Chatbot Training: This dataset can help train chatbots or virtual assistants that handle accounting or customer inquiries by understanding different ways of expressing the same transaction details.

Key Features: High Variability: Sentences are structured in numerous ways to simulate natural human language variations. Randomized Data: Names, dates, products, quantities, prices, and payment methods are randomized, ensuring no duplication. Multi-Field Information: Each record contains key sales information essential for accounting and business use cases.

Potential Applications: Use for Named Entity Recognition (NER) tasks. Apply for information extraction challenges. Create pattern recognition models to understand different sentence structures. Test rule-based systems or machine learning models for sales data entry and accounting automation.

License: Ensure that the dataset is appropriately licensed according to your intended use. For general public and research purposes, choose a CC0: Public Domain license, unless specific restrictions apply.

Context

There is a lack of public available datasets on financial services and specially in the emerging mobile money transactions domain. Financial datasets are important to many researchers and in particular to us performing research in the domain of fraud detection. Part of the problem is the intrinsically private nature of financial transactions, that leads to no publicly available datasets.

We present a synthetic dataset generated using the simulator called PaySim as an approach to such a problem. PaySim uses aggregated data from the private dataset to generate a synthetic dataset that resembles the normal operation of transactions and injects malicious behaviour to later evaluate the performance of fraud detection methods.

Content

PaySim simulates mobile money transactions based on a sample of real transactions extracted from one month of financial logs from a mobile money service implemented in an African country. The original logs were provided by a multinational company, who is the provider of the mobile financial service which is currently running in more than 14 countries all around the world.

This synthetic dataset is scaled down 1/4 of the original dataset and it is created just for Kaggle.

NOTE: Transactions which are detected as fraud are cancelled, so for fraud detection these columns (oldbalanceOrg, newbalanceOrig, oldbalanceDest, newbalanceDest ) must not be used.

Headers

This is a sample of 1 row with headers explanation:

1,PAYMENT,1060.31,C429214117,1089.0,28.69,M1591654462,0.0,0.0,0,0

step - maps a unit of time in the real world. In this case 1 step is 1 hour of time. Total steps 744 (30 days simulation).

type - CASH-IN, CASH-OUT, DEBIT, PAYMENT and TRANSFER.

amount - amount of the transaction in local currency.

nameOrig - customer who started the transaction

oldbalanceOrg - initial balance before the transaction

newbalanceOrig - new balance after the transaction.

nameDest - customer who is the recipient of the transaction

oldbalanceDest - initial balance recipient before the transaction. Note that there is not information for customers that start with M (Merchants).

newbalanceDest - new balance recipient after the transaction. Note that there is not information for customers that start with M (Merchants).

isFraud - This is the transactions made by the fraudulent agents inside the simulation. In this specific dataset the fraudulent behavior of the agents aims to profit by taking control or customers accounts and try to empty the funds by transferring to another account and then cashing out of the system.

isFlaggedFraud - The business model aims to control massive transfers from one account to another and flags illegal attempts. An illegal attempt in this dataset is an attempt to transfer more than 200.000 in a single transaction.

Past Research

There are 5 similar files that contain the run of 5 different scenarios. These files are better explained at my PhD thesis chapter 7 (PhD Thesis Available here http://urn.kb.se/resolve?urn=urn:nbn:se:bth-12932.

We ran PaySim several times using random seeds for 744 steps, representing each hour of one month of real time, which matches the original logs. Each run took around 45 minutes on an i7 intel processor with 16GB of RAM. The final result of a run contains approximately 24 million of financial records divided into the 5 types of categories: CASH-IN, CASH-OUT, DEBIT, PAYMENT and TRANSFER.

Acknowledgements

This work is part of the research project ”Scalable resource-efficient systems for big data analytics” funded by the Knowledge Foundation (grant: 20140032) in Sweden.

Please refer to this dataset using the following citations:

PaySim first paper of the simulator:

E. A. Lopez-Rojas , A. Elmir, and S. Axelsson. "PaySim: A financial mobile money simulator for fraud detection". In: The 28th European Modeling and Simulation Symposium-EMSS, Larnaca, Cyprus. 2016

AI Training Data | Annotated Checkout Flows for Retail, Restaurant, and Marketplace Websites Overview

Unlock the next generation of agentic commerce and automated shopping experiences with this comprehensive dataset of meticulously annotated checkout flows, sourced directly from leading retail, restaurant, and marketplace websites. Designed for developers, researchers, and AI labs building large language models (LLMs) and agentic systems capable of online purchasing, this dataset captures the real-world complexity of digital transactions—from cart initiation to final payment.

Key Features

Breadth of Coverage: Over 10,000 unique checkout journeys across hundreds of top e-commerce, food delivery, and service platforms, including but not limited to Walmart, Target, Kroger, Whole Foods, Uber Eats, Instacart, Shopify-powered sites, and more.

Actionable Annotation: Every flow is broken down into granular, step-by-step actions, complete with timestamped events, UI context, form field details, validation logic, and response feedback. Each step includes:

Page state (URL, DOM snapshot, and metadata)

User actions (clicks, taps, text input, dropdown selection, checkbox/radio interactions)

System responses (AJAX calls, error/success messages, cart/price updates)

Authentication and account linking steps where applicable

Payment entry (card, wallet, alternative methods)

Order review and confirmation

Multi-Vertical, Real-World Data: Flows sourced from a wide variety of verticals and real consumer environments, not just demo stores or test accounts. Includes complex cases such as multi-item carts, promo codes, loyalty integration, and split payments.

Structured for Machine Learning: Delivered in standard formats (JSONL, CSV, or your preferred schema), with every event mapped to action types, page features, and expected outcomes. Optional HAR files and raw network request logs provide an extra layer of technical fidelity for action modeling and RLHF pipelines.

Rich Context for LLMs and Agents: Every annotation includes both human-readable and model-consumable descriptions:

“What the user did” (natural language)

“What the system did in response”

“What a successful action should look like”

Error/edge case coverage (invalid forms, OOS, address/payment errors)

Privacy-Safe & Compliant: All flows are depersonalized and scrubbed of PII. Sensitive fields (like credit card numbers, user addresses, and login credentials) are replaced with realistic but synthetic data, ensuring compliance with privacy regulations.

Each flow tracks the user journey from cart to payment to confirmation, including:

Adding/removing items

Applying coupons or promo codes

Selecting shipping/delivery options

Account creation, login, or guest checkout

Inputting payment details (card, wallet, Buy Now Pay Later)

Handling validation errors or OOS scenarios

Order review and final placement

Confirmation page capture (including order summary details)

Why This Dataset?

Building LLMs, agentic shopping bots, or e-commerce automation tools demands more than just page screenshots or API logs. You need deeply contextualized, action-oriented data that reflects how real users interact with the complex, ever-changing UIs of digital commerce. Our dataset uniquely captures:

The full intent-action-outcome loop

Dynamic UI changes, modals, validation, and error handling

Nuances of cart modification, bundle pricing, delivery constraints, and multi-vendor checkouts

Mobile vs. desktop variations

Diverse merchant tech stacks (custom, Shopify, Magento, BigCommerce, native apps, etc.)

Use Cases

LLM Fine-Tuning: Teach models to reason through step-by-step transaction flows, infer next-best-actions, and generate robust, context-sensitive prompts for real-world ordering.

Agentic Shopping Bots: Train agents to navigate web/mobile checkouts autonomously, handle edge cases, and complete real purchases on behalf of users.

Action Model & RLHF Training: Provide reinforcement learning pipelines with ground truth “what happens if I do X?” data across hundreds of real merchants.

UI/UX Research & Synthetic User Studies: Identify friction points, bottlenecks, and drop-offs in modern checkout design by replaying flows and testing interventions.

Automated QA & Regression Testing: Use realistic flows as test cases for new features or third-party integrations.

What’s Included

10,000+ annotated checkout flows (retail, restaurant, marketplace)

Step-by-step event logs with metadata, DOM, and network context

Natural language explanations for each step and transition

All flows are depersonalized and privacy-compliant

Example scripts for ingesting, parsing, and analyzing the dataset

Flexible licensing for research or commercial use

Sample Categories Covered

Grocery delivery (Instacart, Walmart, Kroger, Target, etc.)

Restaurant takeout/delivery (Ub...

Payments to suppliers made by City of York Council from April 2011 onwards. Resources are split according to financial years. Date: The date shown is the date the transaction was input to the system, not the payment date. Transaction number: Our internal reference number to enable us to identify an individual transaction. Transaction numbers beginning with CR relate to entries from the creditor system, usually straightforward payments or credit notes. Transaction numbers beginning with J relate to journal entries, which are usually an accounting entry to correct a miscoding error. Amount: All payments shown exclude VAT. Negative amounts relate to credit notes or corrections. Corrections: Miscoding errors may occur, for example the allocation of a payment to an incorrect expense area or expense type. These are usually corrected in the next month. One of the principles of the spending guidance is to make the data available quickly and to reflect how each individual item was originally recorded in the financial system. Therefore since this report includes only one months data it is likely to include some miscoding errors which have not been corrected yet. These corrections will not be back dated so will appear in the next months report. In the month that the correction occurs a credit (negative) amount will show against the incorrect expense area/ expense type and the corresponding payment will show against the correct expense area/expense type. Supplier Name: The name of the supplier or recipient of the payment. Payments to individuals which may contain sensitive information have been redacted. Supplier ID: Our internal reference number to enable us to identify the supplier. Expense Area: The department where the expenditure is incurred. Expense Type: The description of the nature of the spend.

Abstract

In order to elucidate the financial lives of smallholder households and build the evidence base on this important client group, Consultative Group to Assist the Poor (CGAP) of the World Bank launched the year-long Financial Diaries with Smallholder Families (the “Smallholder Diaries”). The study captured the financial and in-kind transactions of 270 households in Tanzania, Pakistan and Mozambique, of which 86 households are in the fertile farmlands of western Tanzania. The sample was drawn from 2 villages in Tanzania. Villages were selected based on their involvement in agriculture, and convenience in reaching them. Between June 2014 and July 2015, enumerators visited sample families every fortnight to conduct comprehensive face-to-face interviews to track all the money flowing into and out of their households.

Geographic coverage

In Tanzania, the Smallholder Diaries sites included two villages located in the region of Mbeya, home to one of the largest farming populations in Tanzania. Mbeya sits within the Southern Agricultural Growth Corridor of Tanzania (SAGCOT), a region known for a productive agroecological climate and an array of crops and livestock. Farmers in the region most commonly produce maize, as well as coffee and tea, rice, potatoes, pyrethrum, and cassava. To explore the diversity within this region, Smallholder Diaries sites were selected in two different districts. The two selected villages exhibit important differences in available economic activities, climate, harvest seasons, crops, and use of agricultural inputs.

Analysis unit

The main unit for data collection for transactions was the household. However, each income source and financial instrument was ascribed to a specific household member during the initial questionnaire. Thus all transactions associated with that instrument or income source are registered under its owner. Similarly, transactions related to expenses were individually attributed to the member who initiated the respective transaction.

There was a small number of cash flows where the interviewer was not able to unambiguously identify the initiating household member. In these cases, the cash flow was recorded as belonging to the entire household (in the dataset the member ID field would be blank).

Analysis can be performed at two different levels of aggregation: a) The household itself b) Individual household members

In our study the household is defined as including those who consistently share financial resources, live together, share the same cooking arrangement, and report to the same household head. This includes babies, children, people who travel for work or school during the week and consider the household to be their main residence. However, the definition does not include people who are currently spending an extended period of time away from the household, including college students, students away at boarding school, military personnel, people in prison, or people who live in the house but maintain completely separate expenses (e.g. roommates, other families).

Universe

Once the villages for the Smallholder Diaries were selected, the research teams used a screening process to help identify a range of families with 5 acres of land or less, diverse income sources, access to agricultural inputs, wealth levels, and crops to participate in the research.

In Tanzania, these eligible households were identified using a participatory rural appraisal wealth-ranking technique. Working with committees of village representatives, the research teams conducted wealth-ranking exercises to assess the relative wealth of households in village hamlets or subareas.

Kind of data

Event/Transaction data [evn]

Sampling procedure

The methodology and sample size of the Smallholder Diaries was designed to generate a rich pool of detailed information and insights on a targeted population. The Smallholder Diaries are not intended to be statistically representative of smallholder families in participating countries.

Total number of households in sample: 93 (Mozambique); 86 (Tanzania); 94 (Pakistan). The sample came was drawn from 3 villages in Mozambique, 2 villages in Tanzania, and 2 villages in Pakistan. Villages were selected based on their involvement in agriculture, and convenience in reaching them.

The research teams used a screening process to help identify a range of families with 5 acres of land or less, diverse income sources, access to agricultural inputs, wealth levels, and crops to participate in the research. In Tanzania, these eligible households were identified using a participatory rural appraisal wealth-ranking technique. Working with committees of village representatives, the research teams conducted wealth-ranking exercises to assess the relative wealth of households in village hamlets or subareas.

Mode of data collection

Face-to-face [f2f]

Research instrument

Interviewers visited each household and conducted three initial questionnaires. They 1) collected a household roster and demographic information about household members; 2) captured a register of physical assets and income sources for each household member and 3) registered the unique financial instruments used by each household member. This baseline information was then used to generate a custom cash flows questionnaire for each household, built to collect income, expenditure, and financial transactions for each individual. This customized cash flows questionnaire was then used for the collection of cash flows data. During regular visits about every two weeks, interviewers captured a complete set of daily, individual transactions from the preceding two-week period. Households were asked only about transactions using financial instruments and income sources that they actually have, rather than going through a generic list of questions. However, the cash flows questionnaire was continuously updated as new members joined the household, members acquired new financial instruments or income sources, or as the interviewers became aware of previously undisclosed ones.

Cleaning operations

All data editing was done manually.

Response rate

The sample initially included 286 households in all three countries, and the study ended with 273 households in total – an attrition rate similar to what has been observed in the past in similar Financial Diaries exercises. Households left the study due to moving from the study villages, seasonal migration, and occasionally by the prompting of the research team due to concerns about the household’s willingness to be forthcoming about important sources of income.

The Credit Card Transactions Dataset provides detailed records of credit card transactions, including information about transaction times, amounts, and associated personal and merchant details. This dataset has over 1.85M rows.

How This Dataset Can Be Used:

Fraud Detection : Use machine learning models to identify fraudulent transactions by examining patterns in transaction amounts, locations, and user profiles. Enhancing fraud detection systems becomes feasible by analyzing behavioral patterns.

Customer Segmentation : Segment customers based on spending patterns, location, and demographics. Tailor marketing strategies and personalized offers to these different customer segments for better engagement.

Transaction Classification : Classify transactions into categories such as grocery or entertainment to understand spending behaviors. This helps in improving recommendation systems by identifying transaction categories and preferences.

Geospatial Analysis : Analyze transaction data geographically to map spending patterns and detect regional trends or anomalies based on latitude and longitude.

Predictive Modeling : Build models to forecast future spending behavior using historical transaction data. Predict potential fraudulent activities and financial trends.

Behavioral Analysis : Examine how factors like transaction amount, merchant type, and time influence spending behavior. Study the relationships between user demographics and transaction patterns.

Anomaly Detection : Identify unusual transaction patterns that deviate from normal behavior to detect potential fraud early. Employ anomaly detection techniques to spot outliers and suspicious activities.

These data will appear in [1]. The abstract for that paper is given below: We report on the design, fabrication, and measurement of a Very High Frequency band Josephson Arbitrary Waveform Synthesizer (VHF-JAWS) at frequencies from 1~kHz to 50.05~MHz. The VHF-JAWS chip is composed of a series array of 12,810 Josephson junctions (JJs) embedded in a superconducting coplanar waveguide. Each JJ responds to a pattern of current pulses by creating a corresponding pattern of voltage pulses, each with a time-integrated area related to fundamental constants as $\textit{\textbf{h/2e}}$. The pulse patterns are chosen to produce quantum-based single-tone voltage waveforms with an open-circuit voltage of 50~mV~rms (\mbox{-19.03~dBm} output power into 50~$\Omega$ load impedances) at frequencies up to 50.05~MHz, which is more than twice the voltage that has been generated by previous RF-JAWS designs at 1~GHz. The VHF-JAWS is "quantum-locked", that is, it generates one quantized output voltage pulse per input current pulse per JJ while varying the dc current through the JJ array by at least 0.4~mA and the amplitude of the bias pulses by at least 10~\%. We use the large bias pulse quantum-locking range to investigate one source of error in detail: the direct feedthrough of the current bias pulses into the DUT at VHF frequencies. We reduce this error by high-pass filtering the current bias pulses and measure the error as a function of input pulse amplitude using two techniques: by measuring small changes over the quantum-locking range and by passively attenuating the input pulse amplitude so that the nonlinear JJs no longer generate voltage pulses while the error is only linearly scaled.

Abstract

In order to elucidate the financial lives of smallholder households and build the evidence base on this important client group, Consultative Group to Assist the Poor (CGAP) of the World Bank launched the year-long Financial Diaries with Smallholder Families (the "Smallholder Diaries"). The study captured the financial and in-kind transactions of 270 households in Tanzania, Pakistan and Mozambique, of which 93 households are in impoverished northern Mozambique. The sample came was drawn from 3 villages in Mozambique. Villages were selected based on their involvement in agriculture, and convenience in reaching them. Between June 2014 and July 2015, enumerators visited sample families every fortnight to conduct comprehensive face-to-face interviews to track all the money flowing into and out of their households.

Geographic coverage

In Mozambique, three villages in the Rapale district of northern Nampula Province were selected based on strong recommendations from local stakeholders. While some large companies buy cash crops in the province, smallholders tend to practice the subsistence, rain-fed agriculture that is more commonly found throughout Mozambique.

Analysis unit

The main unit for data collection for transactions was the household. However, each income source and financial instrument was ascribed to a specific household member during the initial questionnaire. Thus all transactions associated with that instrument or income source are registered under its owner. Similarly, transactions related to expenses were individually attributed to the member who initiated the respective transaction.

There was a small number of cash flows where the interviewer was not able to unambiguously identify the initiating household member. In these cases, the cash flow was recorded as belonging to the entire household (in the dataset the member ID field would be blank).

Analysis can be performed at two different levels of aggregation: a) The household itself b) Individual household members

In our study the household is defined as including those who consistently share financial resources, live together, share the same cooking arrangement, and report to the same household head. This includes babies, children, people who travel for work or school during the week and consider the household to be their main residence. However, the definition does not include people who are currently spending an extended period of time away from the household, including college students, students away at boarding school, military personnel, people in prison, or people who live in the house but maintain completely separate expenses (e.g. roommates, other families).

Universe

Once the villages for the Smallholder Diaries were selected, the research teams used a screening process to help identify a range of families with 5 acres of land or less, diverse income sources, access to agricultural inputs, wealth levels, and crops to participate in the research.

In Mozambique, these eligible households were identified using a participatory rural appraisal wealth-ranking technique. Working with committees of village representatives, the research teams conducted wealth-ranking exercises to assess the relative wealth of households in village hamlets or subareas.

Kind of data

Event/Transaction data [evn]

Sampling procedure

The methodology and sample size of the Smallholder Diaries was designed to generate a rich pool of detailed information and insights on a targeted population. The Smallholder Diaries are not intended to be statistically representative of smallholder families in participating countries.

Total number of households in sample: 93 (Mozambique); 86 (Tanzania); 94 (Pakistan). The sample came was drawn from 3 villages in Mozambique, 2 villages in Tanzania, and 2 villages in Pakistan. Villages were selected based on their involvement in agriculture, and convenience in reaching them.

The research teams used a screening process to help identify a range of families with 5 acres of land or less, diverse income sources, access to agricultural inputs, wealth levels, and crops to participate in the research. In Mozambique, these eligible households were identified using a participatory rural appraisal wealth-ranking technique. Working with committees of village representatives, the research teams conducted wealth-ranking exercises to assess the relative wealth of households in village hamlets or subareas.

Mode of data collection

Face-to-face [f2f]

Research instrument

Interviewers visited each household and conducted three initial questionnaires. They 1) collected a household roster and demographic information about household members; 2) captured a register of physical assets and income sources for each household member and 3) registered the unique financial instruments used by each household member. This baseline information was then used to generate a custom cash flows questionnaire for each household, built to collect income, expenditure, and financial transactions for each individual. This customized cash flows questionnaire was then used for the collection of cash flows data. During regular visits about every two weeks, interviewers captured a complete set of daily, individual transactions from the preceding two-week period. Households were asked only about transactions using financial instruments and income sources that they actually have, rather than going through a generic list of questions. However, the cash flows questionnaire was continuously updated as new members joined the household, members acquired new financial instruments or income sources, or as the interviewers became aware of previously undisclosed ones.

Cleaning operations

All data editing was done manually.

Response rate

The sample initially included 286 households in all three countries, and the study ended with 273 households in total – an attrition rate similar to what has been observed in the past in similar Financial Diaries exercises. Households left the study due to moving from the study villages, seasonal migration, and occasionally by the prompting of the research team due to concerns about the household’s willingness to be forthcoming about important sources of income.

Abstract

In order to elucidate the financial lives of smallholder households and build the evidence base on this important client group, Consultative Group to Assist the Poor (CGAP) of the World Bank launched the year-long Financial Diaries with Smallholder Families (the "Smallholder Diaries"). The study captured the financial and in-kind transactions of 270 households in Tanzania, Pakistan and Mozambique, of which 94 households are in the Punjab province, the breadbasket of Pakistan. The sample was drawn from 2 villages in Pakistan. Villages were selected based on their involvement in agriculture, and convenience in reaching them. Between June 2014 and July 2015, enumerators visited sample families every fortnight to conduct comprehensive face-to-face interviews to track all the money flowing into and out of their households.

Geographic coverage

In Pakistan, the Smallholder Diaries were conducted in Bahawalnagar, southern Punjab, within the country's breadbasket. Rice, wheat, and cotton are commonly grown and typically sold through a network of local commission agents (known as arthis) and village traders. Given the dominance of agricultural middlemen in Pakistan, two villages in the district of Bahawalnagar were selected as representative of an area with relatively looser connections to agricultural value chains and middlemen.

Analysis unit

The main unit for data collection for transactions was the household. However, each income source and financial instrument was ascribed to a specific household member during the initial questionnaire. Thus all transactions associated with that instrument or income source are registered under its owner. Similarly, transactions related to expenses were individually attributed to the member who initiated the respective transaction.

There was a small number of cash flows where the interviewer was not able to unambiguously identify the initiating household member. In these cases, the cash flow was recorded as belonging to the entire household (in the dataset the member ID field would be blank).

Analysis can be performed at two different levels of aggregation: a) The household itself b) Individual household members

In our study the household is defined as including those who consistently share financial resources, live together, share the same cooking arrangement, and report to the same household head. This includes babies, children, people who travel for work or school during the week and consider the household to be their main residence. However, the definition does not include people who are currently spending an extended period of time away from the household, including college students, students away at boarding school, military personnel, people in prison, or people who live in the house but maintain completely separate expenses (e.g. roommates, other families).

Universe

Once the villages for the Smallholder Diaries were selected, the research teams used a screening process to help identify a range of families with 5 acres of land or less, diverse income sources, access to agricultural inputs, wealth levels, and crops to participate in the research.

In Pakistan, the sample was selected using a traditional screener survey with questions related to household demographics, crops and livestock, main income sources, and wealth indicators, administered to all households in the selected villages. As a supplement to this process, village leaders and community representatives were consulted to help ensure local participation and eliminate households with large landholdings.

Kind of data

Event/Transaction data [evn]

Sampling procedure

The methodology and sample size of the Smallholder Diaries was designed to generate a rich pool of detailed information and insights on a targeted population. The Smallholder Diaries are not intended to be statistically representative of smallholder families in participating countries.

Total number of households in sample: 93 (Mozambique); 86 (Tanzania); 94 (Pakistan). The sample came was drawn from 3 villages in Mozambique, 2 villages in Tanzania, and 2 villages in Pakistan. Villages were selected based on their involvement in agriculture, and convenience in reaching them.

The research teams used a screening process to help identify a range of families with 5 acres of land or less, diverse income sources, access to agricultural inputs, wealth levels, and crops to participate in the research. In Pakistan, the sample was selected using a traditional screener survey with questions related to household demographics, crops and livestock, main income sources, and wealth indicators, administered to all households in the selected villages. As a supplement to this process, village leaders and community representatives were consulted to help ensure local participation and eliminate households with large landholdings, harvests per year, use of inputs, and integration with local markets and a variety of families were chosen.

In Pakistan, the sample was selected using a traditional screener survey with questions related to household demographics, crops and livestock, main income sources, and wealth indicators. As a supplement to this process, village leaders and community representatives were consulted to help ensure local ownership and eliminate households with large landholdings.

Mode of data collection

Face-to-face [f2f]

Research instrument

Interviewers visited each household and conducted three initial questionnaires. They 1) collected a household roster and demographic information about household members; 2) captured a register of physical assets and income sources for each household member and 3) registered the unique financial instruments used by each household member. This baseline information was then used to generate a custom cash flows questionnaire for each household, built to collect income, expenditure, and financial transactions for each individual. This customized cash flows questionnaire was then used for the collection of cash flows data. During regular visits about every two weeks, interviewers captured a complete set of daily, individual transactions from the preceding two-week period. Households were asked only about transactions using financial instruments and income sources that they actually have, rather than going through a generic list of questions. However, the cash flows questionnaire was continuously updated as new members joined the household, members acquired new financial instruments or income sources, or as the interviewers became aware of previously undisclosed ones.

Cleaning operations

All data editing was done manually.

Response rate

The sample initially included 286 households in all three countries, and the study ended with 273 households in total – an attrition rate similar to what has been observed in the past in similar Financial Diaries exercises. Households left the study due to moving from the study villages, seasonal migration, and occasionally by the prompting of the research team due to concerns about the household’s willingness to be forthcoming about important sources of income.

Dataset Overview Canonical Raw Data represents the ground-truth layer of DeFi and on-chain intelligence. Each sub-dataset captures data at a distinct level of the EVM execution stack - from the raw blocks to raw contract functions results. Each record includes a deterministic _tracing_id, forming the root lineage reference for all derived BlockDB datasets

Chains and Coverage ETH, BSC, Base, Arbitrum, Unichain, Avalanche, Polygon, Celo, Linea, Optimism (others on request). Full history from chain genesis; reorg-aware real-time ingestion and updates.

Included Datasets

1. BlockDB Canonical Raw Blocks (Lineage-Verified) Canonical block-level data including block hashes, parent relationships, miner addresses, gas parameters, and recomputed receipt roots for integrity verification.

2. BlockDB Canonical Raw Transactions (Lineage-Verified) Full transaction-level coverage across all EVM networks, including sender/receiver, input data, gas details, and other fields.

3. BlockDB Canonical Raw Logs (Lineage-Verified) Log events emitted by contracts, normalized across protocols.

4. BlockDB Discovered Smart Contracts Catalog of deployed smart contracts discovered through emmitted logs.

5. BlockDB Discovered Smart Contract Function Results Structured outputs of executed on-chain function calls (e.g., eth_call, view functions).

Lineage Each record of these datasets includes a deterministic _tracing_id, forming the root lineage reference for all derived BlockDB datasets (swaps, liquidity, and token prices). This ensures verifiable traceability, reproducibility, and proof-of-derivation for every downstream record.

Common Use Cases • Establish a canonical on-chain ground truth across multiple EVM chains • Power downstream datasets (tokens, swaps, liquidity, prices) with verified base-layer inputs • Build data lineage visualizations or chain-state replayers for validation and analytics

Quality • Verifiable lineage: deterministic cryptographic hashes per row • Reorg-aware ingestion: continuity and consistency across forks • Complete historical coverage: from chain genesis to present

Bitcoin is a peer-to-peer electronic payment system that popularized rapidly in recent years. Usually, we need to query the complete history of bitcoin blockchain data to acquire variables of economic meaning. This becomes increasingly difficult now with over 1.6 billion historical transactions on the Bitcoin blockchain. It is thus important to query Bitcoin transaction data in a way that is more efficient and provides economic insights. We apply cohort analysis that interprets bitcoin blockchain data using methods developed for population data in social science. Specifically, we query and process the Bitcoin transaction input and output data within each daily cohort. With this, we then create datasets and visualizations for some key indicators of bitcoin transactions, including the daily lifespan distributions of accumulated spent transaction output (STXO) and the daily age distributions of accumulated unspent transaction output (UTXO). We provide a computationally feasible approach to characterize bitcoin transactions, which paves the way for future studies of economic behaviors in the emerging market of Bitcoin.

Agplace Payment FinTech Market Outlook

As per our latest research conducted in 2025, the Agplace Payment FinTech market size reached USD 4.66 billion globally in 2024. The market is expected to grow at a robust CAGR of 18.3% during the forecast period, reaching a projected value of USD 22.86 billion by 2033. This significant growth trajectory is driven by the increasing digitalization of the agricultural sector, rising adoption of mobile financial services among farmers, and expanding integration of advanced payment technologies across the agri-value chain.

The primary growth driver for the Agplace Payment FinTech market is the accelerating digital transformation within the global agriculture ecosystem. As agricultural stakeholders seek to streamline transactions, reduce inefficiencies, and enhance transparency, the demand for innovative fintech solutions tailored to agricultural needs has surged. The proliferation of smartphones and improved internet connectivity in rural areas have further enabled the adoption of digital payment platforms. These platforms facilitate seamless payments for agricultural inputs, produce sales, and insurance premiums, empowering farmers and agribusinesses to operate more efficiently. Furthermore, government initiatives and subsidies promoting digital payments in agriculture have played a pivotal role in fostering market expansion.

Another critical factor contributing to market growth is the diversification of payment types and the evolution of value-added financial services. The market is witnessing a rapid shift from traditional cash-based transactions to digital wallets, mobile payments, and card-based solutions, driven by the need for secure, real-time, and traceable transactions. Financial technology companies are increasingly offering tailored solutions that address the unique challenges of the agricultural sector, such as seasonal cash flows, fragmented supply chains, and risk management. These solutions not only simplify payments but also enable access to credit, insurance, and other financial products, thereby driving financial inclusion and resilience among smallholder farmers and rural communities.

The integration of Agplace Payment FinTech solutions into broader agricultural supply chains is also fueling market growth. By embedding digital payments within platforms for farm management, input procurement, and produce sales, stakeholders can unlock greater operational efficiency and data-driven decision-making. This integration supports end-to-end visibility and traceability, which are increasingly demanded by consumers and regulators alike. In addition, the entry of global fintech players and strategic partnerships with agri-tech startups are fostering innovation and expanding the reach of advanced payment solutions into emerging markets, further accelerating the market’s upward trajectory.

Regionally, Asia Pacific dominates the Agplace Payment FinTech market, accounting for the largest share in 2024, followed by North America and Europe. The strong performance in Asia Pacific can be attributed to its vast agricultural base, rapid digital adoption, and supportive government policies. North America and Europe are experiencing steady growth, driven by technological advancements and the presence of established fintech ecosystems. Meanwhile, Latin America and the Middle East & Africa are emerging as high-potential markets, propelled by increasing investments in rural digital infrastructure and a growing focus on financial inclusion in agriculture.

Component Analysis

The Agplace Payment FinTech market is segmented by component into software, hardware, and services. Software forms the backbone of digital payment solutions in agriculture, encompassing platforms for transaction processing, digital wallets, payment gateways, and analytics tools. The software segment is witnessing rapid innovation, with providers leveraging artificial intelligence, blockchain, and cloud technologies to deliver secure, scalable, and user-friendly solutions. These advanceme

The dataset contains the European Union's social accounting matrix for 2020. The FIGARO database’s 2022 edition (Eurostat (2022). ESA supply, use and input-output tables) is used to create product-by-product input-output table for the EU, while Eurostat’s data on non-financial transactions (a dataset called nasa_10_nf_tr , Eurostat (2022). Non-financial transactions - annual data) is used to cover the remaining parts of the social accounting matrix.

This research was funded by the grant S-MIP-20-53 from the Research Council of Lithuania.

Market Basket Analysis

Market basket analysis with Apriori algorithm

The retailer wants to target customers with suggestions on itemset that a customer is most likely to purchase .I was given dataset contains data of a retailer; the transaction data provides data around all the transactions that have happened over a period of time. Retailer will use result to grove in his industry and provide for customer suggestions on itemset, we be able increase customer engagement and improve customer experience and identify customer behavior. I will solve this problem with use Association Rules type of unsupervised learning technique that checks for the dependency of one data item on another data item.

Introduction

Association Rule is most used when you are planning to build association in different objects in a set. It works when you are planning to find frequent patterns in a transaction database. It can tell you what items do customers frequently buy together and it allows retailer to identify relationships between the items.

An Example of Association Rules

Assume there are 100 customers, 10 of them bought Computer Mouth, 9 bought Mat for Mouse and 8 bought both of them. - bought Computer Mouth => bought Mat for Mouse - support = P(Mouth & Mat) = 8/100 = 0.08 - confidence = support/P(Mat for Mouse) = 0.08/0.09 = 0.89 - lift = confidence/P(Computer Mouth) = 0.89/0.10 = 8.9 This just simple example. In practice, a rule needs the support of several hundred transactions, before it can be considered statistically significant, and datasets often contain thousands or millions of transactions.

Strategy

• Data Import
• Data Understanding and Exploration
• Transformation of the data – so that is ready to be consumed by the association rules algorithm
• Running association rules
• Exploring the rules generated
• Filtering the generated rules
• Visualization of Rule

Dataset Description

• File name: Assignment-1_Data
• List name: retaildata
• File format: . xlsx
• Number of Row: 522065

• Number of Attributes: 7

  • BillNo: 6-digit number assigned to each transaction. Nominal.
  • Itemname: Product name. Nominal.
  • Quantity: The quantities of each product per transaction. Numeric.
  • Date: The day and time when each transaction was generated. Numeric.
  • Price: Product price. Numeric.
  • CustomerID: 5-digit number assigned to each customer. Nominal.
  • Country: Name of the country where each customer resides. Nominal.

https://user-images.githubusercontent.com/91852182/145270162-fc53e5a3-4ad1-4d06-b0e0-228aabcf6b70.png">

Libraries in R

First, we need to load required libraries. Shortly I describe all libraries.

• arules - Provides the infrastructure for representing, manipulating and analyzing transaction data and patterns (frequent itemsets and association rules).
• arulesViz - Extends package 'arules' with various visualization. techniques for association rules and item-sets. The package also includes several interactive visualizations for rule exploration.
• tidyverse - The tidyverse is an opinionated collection of R packages designed for data science.
• readxl - Read Excel Files in R.
• plyr - Tools for Splitting, Applying and Combining Data.
• ggplot2 - A system for 'declaratively' creating graphics, based on "The Grammar of Graphics". You provide the data, tell 'ggplot2' how to map variables to aesthetics, what graphical primitives to use, and it takes care of the details.
• knitr - Dynamic Report generation in R.
• magrittr- Provides a mechanism for chaining commands with a new forward-pipe operator, %>%. This operator will forward a value, or the result of an expression, into the next function call/expression. There is flexible support for the type of right-hand side expressions.
• dplyr - A fast, consistent tool for working with data frame like objects, both in memory and out of memory.
• tidyverse - This package is designed to make it easy to install and load multiple 'tidyverse' packages in a single step.

https://user-images.githubusercontent.com/91852182/145270210-49c8e1aa-9753-431b-a8d5-99601bc76cb5.png">

Data Pre-processing

Next, we need to upload Assignment-1_Data. xlsx to R to read the dataset.Now we can see our data in R.

https://user-images.githubusercontent.com/91852182/145270229-514f0983-3bbb-4cd3-be64-980e92656a02.png"> https://user-images.githubusercontent.com/91852182/145270251-6f6f6472-8817-435c-a995-9bc4bfef10d1.png">

After we will clear our data frame, will remove missing values.

https://user-images.githubusercontent.com/91852182/145270286-05854e1a-2b6c-490e-ab30-9e99e731eacb.png">

To apply Association Rule mining, we need to convert dataframe into transaction data to make all items that are bought together in one invoice will be in ...

This comprehensive dataset offers a thorough and meticulous analysis of Dash transactions, providing a detailed and all-encompassing view. It delves into crucial metrics such as transaction volume, fees, and the overall activity of the network, shedding light on the pulse of the cryptocurrency world. The daily updates not only reflect the dynamic nature of this digital landscape but also make this dataset an essential tool for a diverse range of individuals. Whether you're an astute financial expert conducting in-depth market analyses, a curious researcher unraveling the complexities of the blockchain, or simply a passionate cryptocurrency enthusiast eager to stay informed, this dataset caters to your needs.

If you require further insights or have any inquiries regarding this dataset, please don't hesitate to contact us at info@blockchair.com. Our team is dedicated to assisting you and ensuring you maximize the value of the information provided.

Adaptive Search Ranking AI Market Outlook

The global Adaptive Search Ranking AI market size reached USD 3.2 billion in 2024, according to our latest research, and is expected to grow at a robust CAGR of 24.7% between 2025 and 2033. By the end of the forecast period, the market is projected to achieve a value of USD 27.3 billion. This remarkable growth trajectory is primarily driven by the escalating demand for hyper-personalized search experiences, the proliferation of unstructured data, and the transformative impact of AI-powered search technologies across diverse industries.

One of the primary growth factors for the Adaptive Search Ranking AI market is the exponential rise in digital content and e-commerce transactions. As consumers and enterprises generate vast amounts of data daily, the need for intelligent search solutions that can dynamically adapt to user intent and context has become more urgent than ever. Businesses are increasingly leveraging adaptive search ranking AI to optimize product discovery, streamline information retrieval, and enhance user engagement. This surge in adoption is further accelerated by the growing sophistication of natural language processing (NLP) and machine learning algorithms, which enable search engines to interpret complex queries and deliver highly relevant results in real time.

Another significant driver is the shift towards omnichannel customer experiences and the integration of AI across digital touchpoints. Organizations in sectors such as retail, BFSI, and media are investing in adaptive search ranking AI to unify search experiences across web, mobile, and in-app platforms. This technology allows companies to provide consistent, context-aware recommendations and search outcomes tailored to individual usersÂ’ preferences, browsing history, and behavioral patterns. Additionally, the increasing adoption of voice search, conversational AI, and visual search interfaces is fueling the demand for adaptive search ranking models that can handle multimodal inputs and deliver seamless, intuitive search experiences.

The expansion of cloud infrastructure and the availability of scalable AI platforms are also propelling market growth. Cloud-based deployment enables organizations of all sizes to access advanced adaptive search ranking capabilities without the need for significant upfront investment in hardware or data science expertise. As a result, small and medium enterprises (SMEs) are increasingly embracing these solutions to compete with larger players. Furthermore, advancements in AI model training, real-time data processing, and integration with enterprise systems are making adaptive search ranking AI more accessible and effective, thus widening its adoption across industries.

From a regional perspective, North America continues to dominate the Adaptive Search Ranking AI market owing to its mature technology ecosystem, high digital adoption rates, and the presence of leading AI innovators. However, Asia Pacific is emerging as a high-growth region, driven by rapid digitalization, expanding e-commerce, and increasing investments in AI research and development. Europe also demonstrates strong potential, particularly in sectors such as BFSI and healthcare, where regulatory compliance and data privacy are critical. Latin America and the Middle East & Africa are gradually catching up, supported by government initiatives and rising awareness of AIÂ’s transformative potential.

The emergence of AI Dataset Search Platform is revolutionizing the way organizations access and utilize data for adaptive search ranking. These platforms provide a centralized repository where datasets can be easily discovered, accessed, and shared across various AI applications. By streamlining the dataset search process, organizations can significantly reduce the time and effort required to find relevant data, thereby accelerating the development and deployment of AI models. This innovation is particularly beneficial in the context of adaptive search ranking AI, where the quality and diversity of training data play a crucial role in enhancing the accuracy and relevance of search results. As more organizations recognize the value of AI Dataset Search Platforms, we can expect a surge in their adoption, further driving the growth of the adaptive search ranking AI market.

Real Estate CMA Software Market size was valued at USD 5.1 Billion in 2024 and is projected to reach USD 8.62 Billion by 2031, growing at a CAGR of 7.1% during the forecasted period 2024 to 2031

Global Real Estate CMA Software Market Drivers

Growing Need for Data-driven Decision-Making: Real estate agents are depending more and more on analytics and data to help them make wise choices. With the use of CMA software, which offers thorough data analysis and insights into comparable sales, market trends, and property values, agents and brokers can more successfully negotiate transactions, set listing prices, and evaluate properties with accuracy.

Requirement for a Competitive Advantage: In the current competitive real estate market, brokerages and agents look for solutions that set them apart from rivals and improve the value they offer to clients. Agents can dazzle customers and acquire more listings by using the sophisticated features of CMA software to create professional-looking comparative market assessments, customisable presentations, and interactive reports.

Growing Significance of Engaging Clients: Gaining trust, cultivating relationships, and closing deals in the real estate sector depend on offering clients individualized and engaging experiences. Through visually appealing presentations, interactive maps, and dynamic charts that provide market data and property information in an engaging and understandable manner, agents may effectively engage clients with the help of CMA software.

Simplifying the Listing Presentation Process: Real estate marketing and client acquisition heavily depend on the preparation and delivery of listing presentations. With the help of CMA software, agents can rapidly create professional-looking reports, add branding elements, and show prospective sellers the features, amenities, and market comparisons of their properties. The process of making bespoke listing presentations is also made more efficient and automated.

Integration with Various Data Sources: To obtain thorough and current market data, CMA software integrates with a variety of data sources, such as MLS (Multiple Listing Service) databases, property tax records, public documents, and third-party data providers. The accuracy and reliability of CMAs are increased by this integration, which gives agents access to reliable property information, historical sales data, area demographics, and market statistics.

Efficiency and Time Savings: CMA software saves agents time and effort while creating market studies by automating repetitive operations including data collecting, analysis, and report preparation. CMA software increases efficiency by optimizing workflow procedures and decreasing manual data input, freeing up agents to concentrate more on interacting with clients, generating leads, and completing sales.

Use of sophisticated Technologies: The real estate sector is changing as a result of the use of sophisticated technologies including machine learning (ML), artificial intelligence (AI), and predictive analytics. CMA software helps agents predict market trends, pricing swings, and changes in property worth by using AI and ML algorithms to scan massive information, spot patterns, and produce predictive insights.

Remote Work and Virtual Collaboration: The COVID-19 epidemic has hastened the trend toward remote work and virtual collaboration, which has raised demand for digital solutions that facilitate communication and cooperation from a distance. Agents can make virtual listing presentations, electronically communicate information with clients, and work in real-time team collaborations regardless of their physical locations thanks to CMA software.

Accuracy and Regulatory Compliance: Real estate transactions must adhere to a number of rules and regulations, such as ethical norms, disclosure legislation, and fair housing laws. By offering precise and impartial market evaluations and assisting agents in avoiding the possible legal ramifications of overpricing or underpricing properties, CMA software helps them maintain compliance.

Globalization and Market Expansion: The need for CMA software with international capabilities is driven by the growth of real estate brokerages into new geographic areas and the globalization of real estate markets. Agents can serve clients in a variety of global marketplaces thanks to multilingual support, currency conversion, and localization tools, which facilitate cross-border transactions and global expansion strategies.

The "Daily Transactions" dataset contains information on dummy transactions made by an individual on a daily basis. The dataset includes data on the products that were purchased, the amount spent on each product, the date and time of each transaction, the payment mode of each transaction, and the source of each record (Expense/Income).

This dataset can be used to analyze purchasing behavior and money management, forecasting expenses, and optimizing savings and budgeting strategies. The dataset is well-suited for data analysis and machine learning applications,it can be used to train predictive models and make data-driven decisions.

Column Descriptors

• Date: The date and time when the transaction was made
• Mode: The payment mode used for the transaction
• Category: Each record is divided into a set of categories of transactions
• Subcategory: Categories are further broken down into Subcategories of transactions
• Note: A brief description of the transaction made
• Amount: The transactional amount
• Income/Expense: The indicator of each transaction representing either expense or income
• Currency: All transactions are recorded in official currency of India

Interprovincial Trade Flows (15F0002XDB) The interprovincial and international trade flows for goods and services by province and territory are available at the S-level of commodity aggregation in EXCEL files. National Input - Output Tables (15F0041XDB) The Input-Output accounting system consists of three tables. The input tables (USE tables) detail the commodities that are consumed by various industries. Output tables (MAKE tables) detail the commodities that are prod uced by various industries. Final demand tables detail the commodities bought by many categories of buyers (consumers, industries and government) for both consumption and investment purposes. These tables allow users to track intersectional exchanges of goods and services between industries and final demand categories such as personal expenditures, capital expenditures and public sector expenditures. There are four levels of detail: the "W" or Worksheet level with 303 industries, 727 commodities and 170 final demand categories, the "L" or Link level (the most detailed level that allows the construction of consistent time series of annual data from 1961 to 2002) with 117 industries, 469 commodities and 123 final demand categories, the "M" or Medium level with 62 industries, 111 commodities and 39 final demand categories, and the "S" or Small level with 25 industries, 59 commodities a nd 16 final demand categories. In 2009, several changes were made to the accounting system: there is a new level "D" that is the Detailed level, there are no "M" or "W" level tables, and there are two "L" level tables representing 1961 and 1997 aggregations. Provincial Input-Output Tables (15F0042XDB) The provincial input-output tables are constructed every year. The tables are available at the "S" level only. National and Provincial Multipliers (15F0046XDB) These are a series of Input-Output multipliers and ratios that allow users to quickly estimate the direct, indirect and total impacts of increases in industrial output or increases in an industry's labour force. These are the GDP, labour income, employment and gross output multipliers and ratios. Capital income multipliers and ratios can be calculated by subtracting the labour income figures from the GDP figures. National Symmetric Input-Output Tables - Aggregation Level S (15-207-XC B) The Industry Accounts Division of Statistics Canada publishes annual supply and use input-output (I-O) tables. While these rectangular, industry by commodity closely reflect actual economic transactions, certain analytical and modeling purposes, however, require symmetric industry-by-industry I-O tables. The symmetric industry by industry table shows the inter-industry transactions, that is, all purchases of an industry from all other industries including expenditures on imports and i nventory withdrawals as well as all expenditures on primary inputs. Similarly, the symmetric final demand table shows all purchases by a final demand category from all other industries, including expenditures on imports and inventory withdrawals as well as all expenditures on indirect taxes. National Symmetric Input-Output Tables - Aggregation Level L (15-208-XCB). The Industry Accounts Division of Statistics Canada publishes annual symmetric industry-by-industry I-O tables at the L level. The symmetric industry by industry table shows the inter-industry transactions, that is, all purchases of an industry from all other industries including expenditures on imports and inventory withdrawals as well as all expenditures on primary inputs. Similarly, the symmetric final demand table shows all purchases by a final demand category from all other industries, including expenditures on imports and inventory withdrawals as well as all expenditures on indirect taxes. Provincial GDP by Industry and Sector, at Basic Prices (15-209-XCB). This product presents estimates of Gross Domestic Product (GDP) by industry, in current dollars, evaluated at basic price for all provinces and territories. These estimates are derived from the provincial Input-Output tables. GDP measures the unduplicated value of production. The GDP by industry estimates are derived using a "value added" approach, that is, the value that a producer adds to their intermediate inputs before generating their own output. This allows not only for the computation of total economic production but also the industrial composition and origin of the economic production. When evaluated at basic prices, an industry's GDP is the sum of its factor incomes (wages and salaries, supplementary labour income, mixed income and other operating surplus) plus taxes less subsidies on production (labour and capital). Provincial Gross Output by Industry and Sector (15-210-XCB). This product presents estimates of gross output by industry, in current dollars, evaluated at modified basic price for all provinces and territories. These estimates are derived from the provincial Input-Output tables. Gross output...

This comprehensive dataset offers a thorough and meticulous analysis of Dogecoin transactions, providing a detailed and all-encompassing view. It delves into crucial metrics such as transaction volume, fees, and the overall activity of the network, shedding light on the pulse of the cryptocurrency world. The daily updates not only reflect the dynamic nature of this digital landscape but also make this dataset an essential tool for a diverse range of individuals. Whether you're an astute financial expert conducting in-depth market analyses, a curious researcher unraveling the complexities of the blockchain, or simply a passionate cryptocurrency enthusiast eager to stay informed, this dataset caters to your needs.

If you require further insights or have any inquiries regarding this dataset, please don't hesitate to contact us at info@blockchair.com. Our team is dedicated to assisting you and ensuring you maximize the value of the information provided.