A set of best practices and simple approach for a read-write Linked Data architecture, based on HTTP access to web resources that describe their state using the RDF data model. @en

UberJugaad Enhanced SALT Dataset - About Dataset

🏢 What This Dataset Contains

1.9 million ERP transactions + 151,673 business emails + 3,499 supporting documents

This is an enterprise dataset from UberJugaad GmbH, a €14.8B German industrial supplier with realistic business communications, transactions, and documents. Built on SAP's SALT dataset and enhanced with synthetic business communications for AI/ML applications. File, feature, column descriptions and feature descriptions are in this about me section and listed in further detail in summary .md files.

📊 Dataset Structure & Size

Total Dataset Size: ~160 MB

🎯 Key Features & Use Cases

What Makes This Special

• Realistic Business Ecosystem: Complete email threads, documents, and transactions all linked together
• Natural Language Content: 151K emails with authentic business communication patterns
• Linked Data Architecture: All files connected via order numbers and customer IDs
• Time-Synchronized: Chronologically consistent data from 2019-2020
• Discovery-Oriented: Business patterns embedded in content, not pre-labeled

Perfect For

• Email Classification & Routing - Train models on real business communications
• Sentiment & Urgency Detection - Analyze customer emotions and priority levels
• Customer Behavior Analysis - Understand B2B purchasing patterns and personas
• Document Information Extraction - Process invoices, POs, and business reports
• Business Process Mining - Discover workflows from email and transaction data
• Anomaly Detection - Find unusual patterns in communications and orders
• Multi-Modal AI - Combine text analysis with structured ERP data

🏭 Company Profile: UberJugaad GmbH

Industry: Industrial B2B Distribution & Manufacturing
Scale: €14.8B annual revenue, 7,000-8,500 employees
Operations: 153 locations across 203 countries
Customers: 13,155 active customers (139,611 total)
Products: 164,358 SKUs in industrial supplies and components

Business Model

• Distributors (114): €4.0B revenue - Resellers with high-volume automated ordering
• Small Business (2,921): €5.3B revenue - Regular procurement needs
• Manufacturers (193): €1.2B revenue - Consistent production supply requirements
• Service Companies (867): €1.6B revenue - Emergency repair and maintenance
• Dealers (110): €1.7B revenue - Channel partners with large batch orders

📧 Email Corpus Highlights

Communication Types

• Customer Business Emails: Order issues, complaints, urgent requests
• Internal Escalations: Sales team coordination, problem resolution
• Vendor Communications: Supply chain updates, delivery notifications
• Spam/Marketing: Realistic vendor pitches (eye-BM, Mikerosoft parodies)
• HR Announcements: Company policies, holiday notices
• IT Support: Help desk tickets, system issues

Sample Email Thread

Customer → Sales: "Order 0002456789 arrived damaged, production line at risk"
Sales → Logistics: "URGENT: Major delivery failure, customer threatening €2.3M pullout"
Logistics → Customer: "Expedited replacement shipping today, compensation proposal attached"

Sentiment & Urgency Distribution

• Positive: 45,501 emails (30%) - Confirmations, thanks, routine business
• Neutral: 83,004 emails (55%) - Standard communications, updates
• Negative: 23,168 emails (15%) - Complaints, issues, problems
• Urgency Levels: 0-5 scale, with 23% marked as urgent (3+)

🗃️ Document Types

Supporting Documents (3,467)

• Purchase Orders: Customer orders with terms and delivery requirements
• Invoices: Billing documents with payment terms and amounts
• Shipping Notices: Delivery tracking and logistics updates
• Quality Reports: Inspection results and defect analysis
• Credit Memos: Refunds and billing adjustments

Business Documents (32)

• Meeting Agendas: Sales and production planning meetings
• Quality Metrics: Plant performance and KPI reports
• Overdue Reports: Accounts receivable aging analysis
• Vendor Scorecards: Supplier performance evaluations

📈 ERP Transaction Data

Core Tables

• Sales Documents: 412K orders with customer and shipping details
• Sales Items: 1.9...

Dataset with the raw UNIX timestamps for start/stop time points for each sample of each evaluated configuration as well as produced plots.

The related data for "Designing Efficient DNNs via Hardware-Aware Neural Architecture Search and Beyond".

The related data for "You Only Search Once: On Lightweight Differentiable Architecture Search for Resource-Constrained Embedded Platforms".

Graph and download economic data for Total Revenue for Architectural, Engineering, and Related Services, Establishments Subject to Federal Income Tax (REV5413TAXABL144QNSA) from Q3 2006 to Q2 2025 about architecture, engineering, revenue, establishments, tax, federal, income, and USA.

The dataset presented was collated from various online resouces and comprises data on nest information and other variables for 3175 species of passerine birds. This data has been cleaned and is the ones employed in all analyses. We also attach code and files that are needed to reproduce analyses and main graphs.

Location of the main files and folders relevant to the paper:

All R based analyses were done with a combination of a Windows 8 laptop and a Linux Mint 19.2 Cinnamon. The nparLD analyses require R version 2.15.3, and the other analyses were done with R version 3.4.4. Multiple versions of R can be installed and run on the same system using the Rbuild and Renv packages. Please see the parts of the Rmd files describing the nparLD analyses to see how this was done.

Silk investment and topology related analyses are in the folder 'web-construction FINAL round- raw data/analysis-FINALround/analysis-results/feb_2016_Rmdfiles/

R Markdown notebook files and associated HTML outputs used to replicate figures and results in paper:

Pore-size and coordination number analysis

• beleyur_et_al_topology_analyses.Rmd

Silk investment and per-capita analyses

• beleyur_et_al_silk_analyses.Rmd

Weight-loss analysis

• weight_loss_analysis.Rmd

nparLD analysis scripts

• poresize_CN_nparLD.R
• silk_nparLD.R

Coordination number and pore size calculations

Code and data used to quantify the coordination number and pore size in 'AllResults.zip' - a combination of raw images, .mat data files and .m files that can be analysed using the MATLAB platform.

All data and code in 'AllResults.zip'

Raw data location:

'web-construction FINAL round- raw data/pics-till 11th March/pics by date and batch/'

All raw JPG images are placed according to their batch (A-J), and group size. Please note that not all webs can be finally analysed as having the labelled group size because despite all efforts spiders somehow managed to move between colonies in some cases! Please refer to the actual colonies used in the manuscript, these webs have definitely been built by the specified number of spiders.

Hand-cropped images of only web with retreat

'web-construction FINAL round- raw data/pics-till 11th March/cropped_photos_in_one_place/'

These are TIF files made after manual cropping of the web pictures. Please see the note above explaining how to choose the webs with valid group sizes.

Architectural, engineering and related services price index (AESPI) by North American Industry Classification System (NAICS). Quarterly data are available from the first quarter of 2013. The table includes data for the most recent reference period and the last four periods. The base period for the index is (2018=100).

Graph and download economic data for All Employees: Architectural, Engineering, and Related Services in Michigan (SMU26000006054130001A) from 1990 to 2024 about architecture, engineering, MI, services, employment, and USA.

The related data for "EdgeNAS: Discovering Efficient Neural Architectures for Edge Systems".

Low-Probability-Intercept Data Link Market Outlook

According to our latest research, the global Low-Probability-Intercept (LPI) Data Link market size reached USD 2.17 billion in 2024, demonstrating robust adoption across defense and security sectors. The market is anticipated to grow at a CAGR of 8.6% from 2025 to 2033, culminating in a forecasted value of USD 4.53 billion by 2033. This impressive growth trajectory is primarily driven by escalating demand for secure, resilient communication systems in military and critical infrastructure operations, as well as ongoing technological advancements in signal processing and encryption methodologies.

A primary growth driver for the Low-Probability-Intercept Data Link market is the intensifying focus on secure communications within military and defense operations globally. As modern warfare increasingly leverages electronic and cyber warfare capabilities, the need to protect data links from interception and jamming has become paramount. LPI data links utilize advanced spread spectrum, frequency hopping, and low-power transmission techniques to minimize the probability of detection by adversaries. The proliferation of unmanned aerial vehicles (UAVs), autonomous systems, and network-centric warfare doctrines further amplifies the necessity for robust LPI communications, with defense agencies investing heavily in R&D and procurement of next-generation solutions that ensure operational security in contested environments.

Technological innovation is another significant catalyst propelling the LPI Data Link market forward. The integration of artificial intelligence (AI) and machine learning algorithms into signal processing has led to the development of adaptive and resilient communication protocols capable of dynamically adjusting transmission parameters in real time. These advancements not only enhance the stealth characteristics of LPI systems but also improve bandwidth efficiency and data throughput, meeting the stringent requirements of modern military and aerospace applications. Moreover, the convergence of LPI technologies with emerging 5G and satellite communication infrastructures is opening new avenues for secure, high-speed data links across diverse platforms, including ground vehicles, naval vessels, and space assets.

The evolving threat landscape, marked by the rise of sophisticated electronic surveillance and cyber-attack capabilities, further underscores the critical importance of LPI Data Links. Homeland security agencies, law enforcement, and commercial enterprises with high-value assets are increasingly adopting LPI solutions to safeguard sensitive information and ensure mission continuity. The market is also benefiting from a growing emphasis on interoperability and network-centric operations, which necessitate secure, resilient, and scalable data link architectures. As governments worldwide prioritize information assurance and resilient command-and-control frameworks, the adoption of advanced LPI technologies is expected to accelerate, driving sustained market expansion through 2033.

In the context of modern military operations, the concept of a Low Power Wide Area Battlefield Network is gaining traction as a critical component for maintaining secure and efficient communications. This network architecture is designed to support a wide range of military applications, from logistics and supply chain management to real-time battlefield intelligence and reconnaissance. By leveraging low-power, wide-area technologies, military forces can establish robust, reliable communication links that are less susceptible to detection and interference. This is particularly important in contested environments where traditional communication systems may be compromised. The integration of such networks into existing military infrastructure not only enhances operational efficiency but also ensures that critical data is transmitted securely across vast distances, thereby maintaining the integrity and confidentiality of mission-critical information.

From a regional perspective, North America remains the dominant market for Low-Probability-Intercept Data Links, accounting for the largest revenue share in 2024. This leadership is attributed to substantial defense spending, a mature technological ecosystem, and the presence of leading industry players. However, the Asia

11 figures of this paper. Figure 1 is the hourglass model of the Internet architecture. Figure 2 shows the merge of three hourglasses (data-infrastructure, tools-infrastructure and compute-infrastructure) into the image of a propeller with three blades and the underlying infrastructure. The narrow waist of the hourglass (minimal essential standards and protocols) is comparable to the center of this picture. Figure 3 is the simple Digital Object picture. The smallest conceivable Digital Object is a persistent identifier (PID) (a digital symbol referring to a particular concept). Each digital object that contains “information” should be adorned with metadata asserting things about the nature of that information. Typical intrinsic metadata describe the factual information that is “indisputable” about the digital object itself. Intrinsic metadata containers, expanded metadata containers and the actual containers holding the data elements or the core (in case of for instance a workflow) could also be treated as separate but permanently-linked digital objects, each with their own unique, persistent and resolvable identifier (UPRI) and thus form a stack of related metadata containers that contain (machine readable, FAIR) metadata of different nature, all asserting, however, relevant information about the data container. Figure 4 shows how in the developing Internet of FAIR Data and Services, a linked-data-compliant query in a virtual machine format could automatically find the most relevant databases. Figure 5 shows the semiotic triangle, based on the concept of cancer. Figure 6 shows the single meaningful assertion in machine readable format, which is called a nanopublication. The smallest conceivable assertion has the structure of a subject, a predicate and an object. To form a nanopublication this “triple” needs to be published in machine readable format with full provenance and publication information (also in machine readable format). Figure 7 shows the Knowlet as a collection of cardinal assertions “about” a given subject. The objects effectively form the “conceptual context” of explicitly associated concepts. The predicates can range from very specific and explicit relationship descriptions such as “inhibits” or “is married to” to more generic and less explicit connections, such as “co-occurs in the same sentence as”. Figure 8 shows that the Knowlet is a digital object and needs to be findable, accessible, interoperable and reusable (i.e., FAIR) in its own right. It also may change over time, when more assertions are collected about the core concept. Therefore, each Knowlet in the Internet of FAIR Data and Services (IFDS) needs a unique, persistent and resolvable identifier (UPRI). Figure 9 shows that the Knowlet can be seen as a metadata container for the concept it represents. It can represent many different things from plain concepts like a gene or a person (ORCID record), to a data set, a data base, a work flow or any other thing in the Internet of Things. Figure 10 shows three ways in which Knowlets can be used to connect dispersed digital objects. In Figure 11, A: Concepts, physical objects or things of different semantic types (and thus also intrinsically meaningless unique, persistent and resolvable identifiers (UPRIs)) can cluster based on contextual similarity without ever being explicitly connected (drug might treat disease). B: Nearly identical concepts that are nevertheless in certain circumstances to be seen as distinct, will automatically cluster as one if the resolution of search or matching is lowered, while they will separate out when the resolution is made higher. C: Conceptual and semantic drift occur.

The related data for "SurgeNAS: A Comprehensive Surgery on Hardware-Aware Differentiable Neural Architecture Search". You may also refer to the following Google Drive link for the pretrained weights if you can not download them here https://drive.google.com/drive/folders/1f5R6mNcLoz11eBu8H78feHE8RFiNUHDx?usp=sharing.

Source data associated with Cell-lysis sensing drives biofilm formation in Vibrio cholerae

Article abstract: The ability to associate neutral stimuli with valence information and to store these associations as memories forms the basis for decision making. To determine the underlying computational principles, we build a realistic computational model of a central decision module within the Drosophila mushroom body (MB), the fly’s center for learning and memory. Our model combines the electron microscopy-based architecture of one MB output neuron (MBON-alpha 3), the synaptic connectivity of its 948 presynaptic Kenyon cells (KCs), and its in vivo membrane properties obtained from patch-clamp recordings. We show that this neuron is electrotonically compact and that synaptic input corresponding to simulated odor input robustly drives its spiking behavior. Therefore, sparse innervation by KCs can efficiently control and modulate MBON activity in response to learning with minimal requirements on the specificity of synaptic localization. This architecture allows efficient storage of large numbers of memories using the flexible stochastic connectivity of the circuit. The data-set contains the coding environment and scripts for running the computational model, experimental data for model fitting and all resulting data from computational simulations related to the above mentioned manuscript. The Fitting.zip file stores the resulting trace from fitting the model in tabular form. The Raw Experimental Traces for Fitting.zip file stores the experimental traces that were used for model fitting in tabular form. One .csv file per cell with the 35 traces and the corresponding time. The Simulation_Code.zip file contains the coding environment, simulation scripts and vector data for synapses and model as .py, .swc and .json files. The thirteen Simulation .zip files store all the raw data in a folder resulting from one specific simulation (.dat and .csv files) in tabular form . A subfolder contains .tif, .pdf, .ps files, for the visualization of the plotted data and in some cases example images of activated regions in the dendritic tree of the neuron. For example: Simulation 1 - 1 KC Synapses.zip includes all the data resulting from simulated successive activation of each individual KC. Each trial involved the activation of 10 synapses per KC. amp_PN/soma.dat contains the resulting amplitude of changes in membrane potential at the proximal neurite/soma in mV. num_alphas.csv contains a matrix of the number of activated synapses per trial. PN/soma_voltage.csv contains the resulting changes in membrane potential at the proximal neurite/soma over time in mV. synapse_sites.csv contains the specified numbers of each activated synaptic site during the trial. time.csv contains the time axis for each trial in ms. The subfolder Sim1_Figs contains .pdf and .ps of two example images of the activated synapses from trial 1 and trial 948. Other folders also contain files with plots of the corresponding data.

These data were generated related to the study of the Genetic architecture of transmission stage production and virulence in schistosome parasites.

Abstract: Both theory and experimental data from multiple pathogens suggest that the production of transmission stages should be strongly associated with virulence, but the genetic bases of parasite transmission/virulence traits are poorly understood. In the blood fluke Schistosoma mansoni, parasite genotypes show extensive variation in numbers of cercariae larvae shed from infected snails. Furthermore, high shedding parasites cause high mortality to snails while low shedding parasites cause low mortality, consistent with expected trade-offs between parasite transmission and virulence. To understand the genetic basis of transmission stage production/virulence, we conducted reciprocal crosses between schistosomes from two laboratory populations that differ 8-fold in cercarial shedding and in their virulence to inbred snail hosts. Each parasite generation, we determined four-week cercarial shedding profiles in inbred Biomphalaria glabrata snails infected with single parasite larvae. We sequenced the whole genome of the F0 parents and the exome of the F1 progeny and 188 F2 progeny from each cross, and used linkage mapping to reveal quantitative trait loci (QTLs) underlying transmission stage production. Cercarial production is polygenic: we found three major QTLs on chromosome 1, 3 and 5 (Log-of-the-odds (LOD) = 5.61, 8.19, 6.25) and two minor QTLs on chromosome 2 and 4. These QTLs act additively and explained 28.56% of the phenotypic variation in cercarial shedding. Alleles inherited from the high and low shedding parents were co-dominant at all QTLs, except for chr. 1 and chr. 4 where the “high cercarial shedding” allele is recessive. These results demonstrate that the genetic architecture of key traits directly relevant to schistosome ecology can be dissected using classical linkage mapping approaches, and set the stage for fine mapping and functional validation of the genes involved using the growing armory of functional and cell biology tools available for this parasite.

This dataset is made of 4 tables:

F0_parental_populations.csv

F1.csv

F2.csv

sex.tsv

F0_parental_populations.csv

This table contains the number of cercariae produced by each individual Biomphalaria glabrata Bg26 snails infected with single genotypes of Schistosoma mansoni parasite. We have compared the transmission stage production between two different populations of S. mansoni parasite. This dataset was originally published in Le Clec'h et al., 2019 (Striking differences in virulence, transmission and sporocyst growth dynamics between two schistosome populations. Parasites and Vectors. 2019 Oct 16;12(1):485. doi: 10.1186/s13071-019-3741-z).

This table is made of 9 columns:

id: the unique identifier of each sample.

schistosoma_population: the population of schistosome used for the infection of the snail. Each snail was infected with a single parasite genotype. We have used SmLE (high shedder/highly virulent population) and SmBRE (low shedding/low virulent population).

Shed.1: the number of cercariae produced by each parasite genotype at the first shedding week (4 weeks after exposure to parasite).

Shed.2: the number of cercariae produced by each parasite genotype at the second shedding week (5 weeks after exposure to parasite).

Shed.3: the number of cercariae produced by each parasite genotype at the third shedding week (6 weeks after exposure to parasite).

Shed.4: the number of cercariae produced by each parasite genotype at the fourth shedding week (7 weeks after exposure to parasite).

sum: the sum of the cercariae produced by each parasite genotype over the 4 weeks of shedding (Shed.1 + Shed.2 + Shed.3 + Shed.4).

average: the average number of cercariae produced by each parasite genotype over the 4 weeks of shedding.

sex: the sex of each parasite genotype determined by PCR 1.

F1.csv

This table contains the number of cercariae produced by each individual Biomphalaria glabrata Bg26 snails infected with single genotypes of F1 progeny from the cross SmLE x SmBRE (see the manuscript for details).

This table is made of 11 columns:

id: the unique identifier of each sample.

cross: F1A or F1B cross. Each snail was infected with a single parasite genotype from either F1A or F1B progeny.

Shed.1: the number of cercariae produced by each parasite genotype at the first shedding week (4 weeks after exposure to parasite).

Shed.2: the number of cercariae produced by each parasite genotype at the second shedding week (5 weeks after exposure to parasite).

Shed.3: the number of cercariae produced by each parasite genotype at the third shedding week (6 weeks after exposure to parasite).

Shed.4: the number of cercariae produced by each parasite genotype at the fourth shedding week (7 weeks after exposure to parasite).

sum: the sum of the cercariae produced by each parasite genotype over the 4 weeks of shedding (Shed.1 + Shed.2 + Shed.3 + Shed.4).

average: the average number of cercariae produced by each parasite genotype over the 4 weeks of shedding.

PO: the total phenoloxidase activity in infected snail hemolymph, measured at 7.5 weeks post-exposure 2.

Hb: the hemoglobin rate in infected snail hemolymph, measured at 7.5 weeks post-exposure 3.

sex: the sex of each parasite genotype determined by PCR 1.

F2.csv

This table contains the number of cercariae produced by each individual Biomphalaria glabrata Bg26 snails infected with single genotypes of F2 progeny from the cross SmLE x SmBRE (see the manuscript for details).

This table is made of 10 columns:

id: the unique identifier of each sample.

cross: F2A or F2B cross. Each snail was infected with a single parasite genotype from either F2A or F2B progeny.

Shed.1: the number of cercariae produced by each parasite genotype at the first shedding week (4 weeks after exposure to parasite).

Shed.2: the number of cercariae produced by each parasite genotype at the second shedding week (5 weeks after exposure to parasite).

Shed.3: the number of cercariae produced by each parasite genotype at the third shedding week (6 weeks after exposure to parasite).

Shed.4: the number of cercariae produced by each parasite genotype at the fourth shedding week (7 weeks after exposure to parasite).

sum: the sum of the cercariae produced by each parasite genotype over the 4 weeks of shedding (Shed.1 + Shed.2 + Shed.3 + Shed.4)

average: the average number of cercariae produced by each parasite genotype over the 4 weeks of shedding.

PO: the total phenoloxidase activity in infected snail hemolymph, measured at 7.5 weeks post-exposure 2.

Hb: the hemoglobin rate in infected snail hemolymph, measured at 7.5 weeks post-exposure 3.

sex.csv

This table contains the in silico sexing of F0 parents, F1 parents and F2 progeny of S. mansoni parasites.

This table is made of 4 columns:

id: the unique identifier of each sample

read_depth: the read depth ratio between the Z-linked and pseudo-autosomal regions.

ratio: computed ratio between the Z-linked and pseudo-autosomal regions.

sex: the sex of each parasite genotype determined in silico: a ratio around 1 corresponds to a male carrying two Z chromosomes while a ratio around 0.5 corresponds to a female carrying only one Z chromosome.

Notes:

1. Le Clec’h W, Chevalier F et al. Real-time PCR for sexing Schistosoma mansoni cercariae. Mol Biochem Parasitol. Jan-Feb 2016; 205(1-2):35-8.doi: 10.1016/j.molbiopara.2016.03.010. Epub 2016 Mar 26.

2. Le Clec’h W et al. Characterization of hemolymph phenoloxidase activity in two Biomphalaria snail species and impact of Schistosoma mansoni infection. Parasit Vectors. 2016 Jan 22; 9:32.doi: 10.1186/s13071-016-1319-6.

3. Le Clec'h et al. Striking differences in virulence, transmission and sporocyst growth dynamics between two schistosome populations. Parasit Vectors. 2019 Oct 16; 12(1):485. doi: 10.1186/s13071-019-3741-z.

Evolution of ecomorphologically relevant traits such as body shapes is important to colonize and persist in a novel environment. Habitat-related adaptive divergence of these traits is therefore common among animals. We studied the genomic architecture of habitat-related divergence in the body shape of Gnathopogon fishes, a novel example of lake–stream ecomorphological divergence, and tested for the action of directional selection on body shape differentiation. Compared to stream-dwelling Gnathopogon elongatus, the sister species Gnathopogon caerulescens, exclusively inhabiting a large ancient lake, had an elongated body, increased proportion of the caudal region and small head, which would be advantageous in the limnetic environment. Using an F2 interspecific cross between the two Gnathopogon species (195 individuals), quantitative trait locus (QTL) analysis with geometric morphometric quantification of body shape and restriction-site associated DNA sequencing-derived markers (1622 loci) identified 26 significant QTLs associated with the interspecific differences of body shape-related traits. These QTLs had small to moderate effects, supporting polygenic inheritance of the body shape-related traits. Each QTL was mostly located on different genomic regions, while colocalized QTLs were detected for some ecomorphologically relevant traits that are proxy of body and caudal peduncle depths, suggesting different degree of modularity among traits. The directions of the body shape QTLs were mostly consistent with the interspecific difference, and QTL sign test suggested a genetic signature of directional selection in the body shape divergence. Thus, we successfully elucidated the genomic architecture underlying the adaptive changes of the quantitative and complex morphological trait in a novel system.

A dataset composed of 8 software engineering projects collected (and refined) from the Software Engineering – Laboratory of Advanced Programming” course at Sapienza University of Rome for master students in Engineering in Computer Science.

The dataset comprises folders for each project. Each folder contains:

• input.txt file with the description of the system and the list of user stories,
• DataMetrics.json with the performance characteristics of the project. An example of DataMetrics format is provided below.
  

  [
   {
    "set_id": 1,
    "set_name": "auth client",
    "user_stories": [1, 2, 3, 4],
    "links": [2, 3, 5],
    "db": "true"
   },
   ...
  ]

  The json is represented by an array of dictionaries, each relative to a set, characterized by a set_id and a set_name, grouping user stories (identified by their numerical identifier in user_stories). Each dictionary also contains links and db keys to indicate other sets that have a related context and the need for a backend service to store or retrieve data, respectively. From an architectural point of view, user stories that belong to linked sets can be fulfilled by the same container and the sets of user stories that are required to store or retrieve data must be fulfilled by a container hosting a database microservice.

• Student_Doc.md with the student (entire) documentation (solution) of the project,
• source code of the developed project.

The dataset is under continuous updating. Each academic year it will be enriched with new projects.

If you want to contribute, please contact us.

Graph and download economic data for All Employees, Architectural, Engineering, and Related Services (CES6054130001) from Jan 1990 to Sep 2025 about architecture, engineering, professional, establishment survey, business, services, employment, and USA.