tasksource/help-desk-tickets
Help Desk Tickets Processed tables derived from version 3 of Mohammad Abdellatif's Mendeley Data dataset. The source contains real helpdesk tickets and associated workflow records from an international software company, covering January 2007 through March 2023. Identifiers and message content were masked by the source authors to protect privacy while retaining context. The default config is reporter_messages, with one row per issue that has reporter-authored text. Utterances are… See the full description on the dataset page: https://huggingface.co/datasets/tasksource/help-desk-tickets.
Help Desk Tickets
Processed tables derived from version 3 of Mohammad Abdellatif's Mendeley Data dataset. The source contains real helpdesk tickets and associated workflow records from an international software company, covering January 2007 through March 2023. Identifiers and message content were masked by the source authors to protect privacy while retaining context.
The default config is `reporter_messages`, with one row per issue that has reporter-authored text. Utterances are kept in chronological order and joined with newline separators. Assignee and other-participant replies are excluded. The issues config provides a compact issue-metadata table.
These are derived data views, not supervised train/dev/test splits. Mendeley v3 does not provide a seven-class category field in issues.csv; its issue_type field is a separate ticket/work-item type. Therefore no classification task is defined here. The reporter-message view contains 357 text-bearing issues; their issue_type values are Ticket (339), HD Service (9), and Deployment (9), not the requested seven-category target.
Columns
Default: reporter_messages
issue_id: source issue identifiertext: chronological concatenation of reporter-authored utterances, with source text otherwise left unchangedissue_type,issue_priority,issue_resolution,issue_statusissue_created,issue_resolution_date,issue_comments_countmessage_count: number of included reporter utterances
issues
One row per source issue with issue_id, issue_num, issue_proj, issue_type, issue_priority, issue_resolution, issue_status, creation and resolution timestamps, comment count, total workflow time, and processing step count.
Source and license
- Original title: Help Desk Tickets
- Author: Mohammad Abdellatif
- DOI: 10.17632/btm76zndnt.3
- License: CC BY 4.0
The source provides PII-masked records and messages. Its language mix includes English and other languages; no language filtering or text correction is applied here.
This mirror uses Mendeley v3 (DOI 10.17632/btm76zndnt.3). The core issue, snapshot, history, utterance, and sample files have identical SHA-256 hashes in v2 and v3; v3 adds holidays_by_country.csv and updated documentation. We do not duplicate the identical v2 data.
Citation
Mohammad Abdellatif. Help Desk Tickets. Mendeley Data, version 3. https://doi.org/10.17632/btm76zndnt.3
