Context

This is complaints ticket data

Content

Complaint texts that belong to various departments of a financial institution

Acknowledgments

We wouldn't be here without the help of others. If you owe any attributions or thanks, include them here and any citations of past research.

Inspiration

We need to cluster the data into various categories using Topic Modelling.

About Dataset

Check out my new Dataset:

Multilingual Support Tickets

Dataset Description:

The Email Ticket Text Multi-Language Classification dataset offers an extensive and meticulously organized collection of email tickets written in multiple languages, including German, English, Spanish, and French. This dataset is specifically designed to enhance natural language processing (NLP), machine learning, and customer support optimization efforts. It provides a rich resource for developing and testing models that classify and prioritize customer support tickets. The dataset's structure and comprehensiveness make it ideal for both academic research and practical applications in improving customer service efficiency.

Dataset Features:

1. Queue:

   • Description: Specifies the department to which the email ticket is categorized. This helps in routing the ticket to the appropriate support team for resolution.
   • Values:
     • Software: Issues related to software applications, systems, bugs, and user errors.
     • Hardware: Problems related to physical devices, hardware malfunctions, installation issues, and maintenance requests.
     • Accounting: Inquiries and issues concerning financial transactions, billing, account discrepancies, and other accounting-related matters.

2. Priority:

   • Description: Indicates the urgency and importance of the issue. Helps in managing the workflow by prioritizing tickets that need immediate attention.
   • Values:
     • 1 (Low): Non-urgent issues that do not require immediate attention. Examples: General inquiries, minor inconveniences, routine updates, and feature requests.
     • 2 (Medium): Moderately urgent issues that need timely resolution but are not critical. Examples: Performance issues, intermittent errors, and user questions requiring detailed responses.
     • 3 (Critical): Urgent issues that require immediate attention and quick resolution. Examples: System outages, security breaches, data loss, and major malfunctions.

3. Software Used:

   • Description: Specifies the software application involved in the issue. Useful for categorizing and analyzing software-related problems.
   • Values: Examples include specific software names like "Arbitrum," "Adobe Premiere Pro 2021," "Excel," etc.

4. Hardware Used:

   • Description: Specifies the hardware device involved in the issue. Helps in identifying and troubleshooting hardware-related problems.
   • Values: Examples include specific hardware names like "Wireless Mouse," "IP PBX," "SFX-Netzteil," etc.

5. Accounting Category:

   • Description: Specifies the sub-category within accounting for more granular classification of financial inquiries.
   • Values: Examples include categories like "Customer Inquiries::Technical Support," "Employee Inquiries::Technical," "Customer Inquiries::Cancellations," etc.

6. Language:

   • Description: Indicates the language in which the email is written. Useful for language-specific NLP models and multilingual support analysis.
   • Values: Examples include "en" (English), "de" (German), "es" (Spanish), "fr" (French).

7. Subject:

   • Description: Provides a brief overview of the email content, aiding in quick scanning and initial categorization.
   • Values: Examples include subject lines like "Wireless Mouse suddenly stops working," "Problème de connexions IP PBX," "Problem mit meinem SFX-Netzteil," etc.

8. Text:

   • Description: Contains the full email text for in-depth analysis. Crucial for training NLP models to understand and classify complex and varied text inputs.
   • Values: Full content of the emails, describing the issue in detail.

Usage Instructions:

To access the full dataset, please contact me at dataset@softoft.de. This page only provides a preview with 200 randomly selected rows from the complete dataset, which comprises over 8,000 rows. More information about me and my company can be found at Softoft. For detailed information about this dataset, visit the dataset website.

Key Features:

• Email Subject: Provides a brief overview of the email content, aiding in quick scanning and initial categorization.
• Email Texts: Contains the full email text for in-depth analysis, crucial for training NLP models.
• Department Classification: Categorizes emails into Software, Hardware, and Accounting departments, enabling specialized handling.
• Priority Levels: Ranks emails based on urgency from low to critical, ensuring that critical issues are addressed promptly.

Use Cases:

1. Text Classification:
   • Tra...

This dataset contains detailed records of IT service tickets, combining structured metadata (such as priority, category, and assignment) with rich ticket descriptions suitable for natural language processing. It enables automated ticket triage, prioritization, and advanced analytics for IT support operations, making it ideal for machine learning and process optimization.

The dataset contains 47,837 rows of data and 2 features.

Features: "Document": ticket text/content "Topic_group": ticket category

Categories include: 'Hardware', 'HR Support', 'Access', 'Miscellaneous', 'Storage', 'Purchase', 'Internal Project', 'Administrative rights'

Collection of 2229 support tickets manually classified into 7 categories, obtained from a IT support company in the Florianópolis (Brazil) region. Each ticket is represented by an unstructured text field, which is typed by the user that opened the call. The classification process was performed in 2020 by three IT support professionals. The corpus contains tickets in many languages, mainly English, German, Portuguese and Spanish.

All Personal Identifiable Information (PII) and sensitive information were removed (substituted by a tag indicating the original content, for instance: the sentence "this text was written by Leonardo" is converted to "this text was written by [NAME]"). The removal was performed in three steps: first, the automated machine learning-based tool AWS Comprehend PII Removal was used; then, a sequence of custom regular expressions was applied; last, the entire corpus was manually verified.

Problem statement

For a financial company, customer complaints carry a lot of importance, as they are often an indicator of the shortcomings in their products and services. If these complaints are resolved efficiently in time, they can bring down customer dissatisfaction to a minimum and retain them with stronger loyalty. This also gives them an idea of how to continuously improve their services to attract more customers. These customer complaints are unstructured text data; so, traditionally, companies need to allocate the task of evaluating and assigning each ticket to the relevant department to multiple support employees. This becomes tedious as the company grows and has a large customer base. In this case study, you will be working as an NLP engineer for a financial company that wants to automate its customer support tickets system. As a financial company, the firm has many products and services such as credit cards, banking and mortgages/loans.

Business goal

You need to build a model that is able to classify customer complaints based on the products/services. By doing so, you can segregate these tickets into their relevant categories and, therefore, help in the quick resolution of the issue. With the help of topic modelling, you will detect patterns and recurring words present in each ticket. This can be then used to understand the important features for each cluster of categories. By segregating the clusters, you will be able to identify the topics of the customer complaints. You will be doing topic modelling on the .json data provided by the company. Since this data is not labelled, you need to apply techniques to analyze patterns and classify tickets into the following five clusters based on their products/services:

1. Credit card / Prepaid card

2. Bank account services

3. Theft/Dispute reporting

4. Mortgages/loans

5. Others

With the help of topic modelling, you will be able to map each ticket onto its respective department/category. You can then use this data to train any supervised model such as logistic regression, decision tree or random forest. Using this trained model, you can classify any new customer complaint support ticket into its relevant department.

Dataset

The data set given to you is in the .json format and contains 78,313 customer complaints with 22 features.

Featuring Labeled Customer Emails and Support Responses

  🔧 Synthetic IT Ticket Generator — Custom Dataset

Create a dataset tailored to your own queues & priorities (no PII). 👉 Generate custom data

Define your queues, priorities, language

Need an on-prem AI to auto-classify tickets?→ Open Ticket AI There are 2 Versions of the dataset, the new version has more tickets, but only languages english and german. So please look at both files, to find what best fits your needs.… See the full description on the dataset page: https://huggingface.co/datasets/Tobi-Bueck/customer-support-tickets.

Overview

Synthetic tabular dataset of 50,000 support tickets from 25 companies used to study priority classification (low, medium, high). Companies differ by size and industry; large companies operate across multiple regions. Features mix numeric and categorical signals commonly available at ticket intake. Data is fully artificial—no real users, systems, or proprietary logs.

Intended use: benchmarking supervised learning for tabular classification (e.g., Gradient Boosting, XGBoost, LightGBM, AdaBoost, SVM, Naive Bayes), feature engineering, handling mixed types, class imbalance, and mild label noise.

File & schema

Identifiers & time

-ticket_id (int64): unique ticket identifier (randomized order)

-day_of_week (Mon–Sun), day_of_week_num (1–7; Mon=1)

Company profile (replicated per row)

-company_id (int), company_size (Small/Medium/Large + _cat),

-industry (7 categories + _cat),

-customer_tier (Basic/Plus/Enterprise + _cat),

-org_users (int): active user seats (Large up to ~10,000)

Context

-region (AMER/EMEA/APAC + _cat)

-past_30d_tickets (int), past_90d_incidents (int)

Product & channel

-product_area (auth, billing, mobile, data_pipeline, analytics, notifications + _cat)

-booking_channel (web, email, chat, phone + _cat)

-reported_by_role (support, devops, product_manager, finance, c_level + _cat)

Impact & flags

-customers_affected (int, heavy-tailed)

-error_rate_pct (float, 0–100; sometimes 0.0 as “unmeasured”)

-downtime_min (int, 0 when only degraded)

-payment_impact_flag, security_incident_flag, data_loss_flag, has_runbook (0/1)

Text proxy

-customer_sentiment (negative/neutral/positive + _cat with 0 = missing)

-description_length (int, 20–2000)

Target

-priority (low/medium/high + priority_cat = 1/2/3)

Notes & limitations

• Fully synthetic; suitable for education, benchmarking, and tutorials.
• No temporal ordering or post-resolution fields are included, avoiding label leakage.
• The noise level is tuned for ~97–98% ceiling performance with well-optimized models, not perfect separability.

This dataset provides comprehensive logs of customer support tickets, including detailed issue descriptions, resolution times, outcomes, agent involvement, and customer satisfaction ratings. It enables analysis of support process efficiency, identification of bottlenecks, and development of NLP models for automated ticket classification and resolution prediction.

Dataset

This dataset was created by Aniket

Contents

The Customer Support Ticket Dataset is a dataset that includes customer support tickets for various tech products. It consists of customer inquiries related to hardware issues, software bugs, network problems, account access, data loss, and other support topics. The dataset provides information about the customer, the product purchased, the ticket type, the ticket channel, the ticket status, and other relevant details.

The dataset can be used for various analysis and modelling tasks in the customer service domain.

Features Description:

• Ticket ID: A unique identifier for each ticket.
• Customer Name: The name of the customer who raised the ticket.
• Customer Email: The email address of the customer (Domain name - @example.com is intentional for user data privacy concern).
• Customer Age: The age of the customer.
• Customer Gender: The gender of the customer.
• Product Purchased: The tech product purchased by the customer.
• Date of Purchase: The date when the product was purchased.
• Ticket Type: The type of ticket (e.g., technical issue, billing inquiry, product inquiry).
• Ticket Subject: The subject/topic of the ticket.
• Ticket Description: The description of the customer's issue or inquiry.
• Ticket Status: The status of the ticket (e.g., open, closed, pending customer response).
• Resolution: The resolution or solution provided for closed tickets.
• Ticket Priority: The priority level assigned to the ticket (e.g., low, medium, high, critical).
• Ticket Channel: The channel through which the ticket was raised (e.g., email, phone, chat, social media).
• First Response Time: The time taken to provide the first response to the customer.
• Time to Resolution: The time taken to resolve the ticket.
• Customer Satisfaction Rating: The customer's satisfaction rating for closed tickets (on a scale of 1 to 5).

Use Cases of such dataset:

• Customer Support Analysis: The dataset can be used to analyze customer support ticket trends, identify common issues, and improve support processes.
• Natural Language Processing (NLP): The ticket descriptions can be used for training NLP models to automate ticket categorization or sentiment analysis.
• Customer Satisfaction Prediction: The dataset can be used to train models to predict customer satisfaction based on ticket information.
• Ticket Resolution Time Prediction: The dataset can be used to build models for predicting the time it takes to resolve a ticket based on various factors.
• Customer Segmentation: The dataset can be used to segment customers based on their ticket types, issues, or satisfaction levels.
• Recommender Systems: The dataset can be used to build recommendation systems for suggesting relevant solutions or products based on customer inquiries.

Overview This dataset comprises detailed records of customer support tickets, providing valuable insights into various aspects of customer service operations. It is designed to aid in the analysis and modeling of customer support processes, offering a wealth of information for data scientists, machine learning practitioners, and business analysts.

Dataset Description The dataset includes the following features:

Ticket ID: Unique identifier for each support ticket. Customer Name: Name of the customer who submitted the ticket. Customer Email: Email address of the customer. Customer Age: Age of the customer. Customer Gender: Gender of the customer. Product Purchased: Product for which the customer has requested support. Date of Purchase: Date when the product was purchased. Ticket Type: Type of support ticket (e.g., Technical Issue, Billing Inquiry). Ticket Subject: Brief subject or title of the ticket. Ticket Description: Detailed description of the issue or inquiry. Ticket Status: Current status of the ticket (e.g., Open, Closed, Pending). Resolution: Description of how the ticket was resolved. Ticket Priority: Priority level of the ticket (e.g., High, Medium, Low). Ticket Channel: The Channel through which the ticket was submitted (e.g., Email, Phone, Web). First Response Time: Time taken for the first response to the ticket. Time to Resolution: Total time taken to resolve the ticket. Customer Satisfaction Rating: Customer satisfaction rating for the support received. Usage This dataset can be utilized for various analytical and modeling purposes, including but not limited to:

Customer Support Analysis: Understand trends and patterns in customer support requests, and analyze ticket volumes, response times, and resolution effectiveness. NLP for Ticket Categorization: Develop natural language processing models to automatically classify tickets based on their content. Customer Satisfaction Prediction: Build predictive models to estimate customer satisfaction based on ticket attributes. Ticket Resolution Time Prediction: Predict the time required to resolve tickets based on historical data. Customer Segmentation: Segment customers based on their support interactions and demographics. Recommender Systems: Develop systems to recommend products or solutions based on past support tickets. Potential Applications: Enhancing customer support workflows by identifying bottlenecks and areas for improvement. Automating the ticket triaging process to ensure timely responses. Improving customer satisfaction through predictive analytics. Personalizing customer support based on segmentation and past interactions. File information: The dataset is provided in CSV format and contains 8470 records and [number of columns] features.

This dataset provides detailed records of telecom customer support tickets, including issue types, resolution timelines, agent actions, and customer satisfaction ratings. It enables process optimization, root cause analysis, and AI/ML chatbot training by offering granular insights into ticket lifecycles and outcomes.

According to a survey conducted in November 2022 in the Kanto region in Japan, both **** percent of male respondents and **** percent of female respondents used traditional paper tickets at least once. Online tickets that need to be issued at convenience stores became the second most-popular option among both men and women, with **** percent and **** percent, respectively.

Here are a few use cases for this project:

1. Expense Management: This model can be used to automate the process of capturing and categorizing data from various expenses such as invoices, receipts, tolls, and travel tickets. Users can simply take a photo of the document and the model will identify and categorize the data, reducing the manual effort and potential for errors.

2. Travel Services: Travel agencies or services could use it to categorize different types of tickets (plane, train, taxi) from images or scanned documents, helping to automate the process of logging and tracking travel data for their customers.

3. Document Archiving and Retrieval: Companies dealing with a large volume of paperwork, such as law firms or accountancy firms, could use this model to categorize, archive and retrieve various documents like invoices, receipts, and certificates more efficiently.

4. Tax Audit: During tax audits, both by individuals or businesses, this model could be used to quickly and accurately organize financial documents including various invoices and receipts.

5. Logistics & Supply Chain: This model can be used to automate the process of tracking shipments or consignments by scanning and recognizing QR codes.

Dataset

This dataset was created by SagarShinde_SagarShinde

Released under Apache 2.0

Contents

The global event ticketing system market is experiencing robust growth, driven by increasing adoption of online ticketing platforms, the rising popularity of diverse events (concerts, sporting events, conferences), and the expanding reach of digital technologies. The market, estimated at $15 billion in 2025, is projected to grow at a Compound Annual Growth Rate (CAGR) of 12% from 2025 to 2033, reaching an estimated market size of approximately $45 billion by 2033. This significant expansion is fueled by several key factors. Firstly, the shift from traditional, on-site ticketing methods to online platforms offers greater convenience, efficiency, and scalability for event organizers and attendees alike. Secondly, the increasing use of mobile ticketing and integrated payment gateways streamlines the purchasing process and enhances the overall user experience. Thirdly, the rise of sophisticated event management software that integrates ticketing functionality allows for better audience management, personalized marketing, and enhanced data analysis. Finally, the continued growth of the events industry itself, encompassing a wider variety of event types and reaching new demographics, provides a strong foundation for continued market expansion. However, challenges remain. Competition among numerous established and emerging players, including giants like Ticketmaster and smaller niche players, necessitates continuous innovation and differentiation. Furthermore, the need to ensure secure payment gateways and data protection remains paramount, as does addressing potential issues like ticket fraud and scalping. The market is segmented by ticketing method (online vs. on-site) and event type (concerts, sporting events, conferences, etc.), each segment displaying unique growth trajectories based on factors such as event popularity, technological adoption, and pricing strategies. Geographical variations are also significant, with mature markets in North America and Europe showing steady growth while developing markets in Asia-Pacific and other regions offer promising expansion opportunities.

Ticket Fraud Detection for Events Market Outlook

According to our latest research, the global Ticket Fraud Detection for Events market size reached USD 2.13 billion in 2024 and is projected to grow at a CAGR of 17.5% from 2025 to 2033, reaching USD 8.32 billion by 2033. The robust growth in this market is primarily driven by the increasing adoption of digital ticketing platforms, rising incidents of ticket fraud, and the need for advanced security measures to ensure seamless event experiences worldwide.

The rapid proliferation of online ticketing systems has significantly contributed to the growth of the Ticket Fraud Detection for Events market. As more consumers shift towards purchasing event tickets through digital channels, the risk of fraudulent activities such as scalping, counterfeit ticket sales, and identity theft has escalated. This trend has compelled event organizers, ticketing platforms, and venue operators to invest heavily in sophisticated fraud detection solutions that leverage AI, machine learning, and blockchain technologies. These advanced tools are capable of analyzing vast volumes of transactional data in real time, identifying suspicious patterns, and preventing unauthorized access to events. The growing awareness among stakeholders regarding the financial and reputational impact of ticket fraud is further fueling the adoption of comprehensive fraud detection systems across the events industry.

Another major growth factor for the Ticket Fraud Detection for Events market is the integration of multi-layered security frameworks. Organizations are increasingly deploying solutions that combine biometric authentication, two-factor verification, and blockchain-based ticketing to enhance the security of their ticketing processes. The rise of hybrid and virtual events post-pandemic has also expanded the threat landscape, necessitating more robust fraud detection mechanisms. In addition, regulatory requirements and industry standards, such as the General Data Protection Regulation (GDPR) and Payment Card Industry Data Security Standard (PCI DSS), have mandated stricter controls over personal and financial data, prompting event organizers to prioritize fraud prevention measures. The continuous evolution of fraud tactics, including the use of bots and deepfakes, is pushing solution providers to innovate and stay ahead of cybercriminals.

The increasing collaboration between event organizers, ticketing platforms, and technology vendors is another crucial factor propelling market growth. Strategic partnerships and integrations are enabling seamless data sharing and more comprehensive fraud detection capabilities. Additionally, the expansion of live events, concerts, sports tournaments, and exhibitions in emerging economies is creating new opportunities for market players. These regions are witnessing a surge in disposable income, urbanization, and digital infrastructure, all of which contribute to higher ticket sales and, consequently, a greater need for fraud detection solutions. However, the high cost of implementing advanced fraud detection systems and the complexity of integrating them with legacy ticketing platforms may pose challenges, particularly for small and medium enterprises.

From a regional perspective, North America continues to dominate the Ticket Fraud Detection for Events market, accounting for the largest revenue share in 2024. This leadership is attributed to the region’s mature event management industry, high digital adoption rates, and the presence of major technology providers. Europe follows closely, driven by strong regulatory frameworks and a vibrant cultural events scene. Meanwhile, the Asia Pacific region is emerging as the fastest-growing market, fueled by rapid urbanization, increasing internet penetration, and a burgeoning middle-class population engaging in live entertainment and sports events. Latin America and the Middle East & Africa are also experiencing steady growth, albeit at a slower pace, due to rising investments in event infrastructure and digital technologies.

Component Analysis

The Ticket Fraud Detection for Events market is segmented by component into software, hardware, and services. The software segment holds the largest market share, primarily due to the increasing demand for advanced analytics, artificial intelligence, and machine learning-powered fraud detection platforms. These software solutions are designed to monitor ticketing transacti

TicketClassify

## Overview

TicketClassify is a dataset for classification tasks - it contains Tickets annotations for 255 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [MIT license](https://creativecommons.org/licenses/MIT).

The most common ticket type in public transportation in Sweden was the season ticket in 2019, with a share of ** percent. For comparison, single tickets accounted for ***** percent of the total ticket sales.