WIT: Workshop On Deriving Insights From User-Generated Text @KDD2021

Workshop Overview

Users from Web platforms and online services generate tremendous amounts of user-generated data in the form of search requests, reviews, questions, and answers. We believe there is tremendous opportunity to exploit advanced AI/NLP techniques on user-generated text data which are rich in user insights and experiences.

The WIT workshop provides a venue for researchers from the academia and industry to address challenges around harnessing text-heavy user-generated data that is available to different types of organizations, especially on topics pertaining to the pipeline of extracting data from unstructured text to a structured form to obtain insights. We will have a great line-up of Invited Speakers and Panelists.  

WIT will be held as a virtual single-day event at SIGKDD 2021


Confirmed Invited Speakers

Confirmed Speaker (alphabetically ordered list):


The Workshop will happen August 15, 2021 (all times are PST).
8:00am – 8:10am: Welcome Message
8:10am – 9:10am: Invited Talk by Luna Dong
9:10am – 9:40am: Session I – Oral  Presentations (2 papers)
  • Evan Shieh, Saul Simhon, Geetha Aluri, Giorgos Papachristoudis, Doa Yakut and Dhanya Raghu. Attribute Similarity and Relevance-Based Product Schema Matching for Targeted Catalog Enrichment
  • Natasha Z. Foutz, Xuan Zhang, Zhilei Qiao, Wenqi Shen and Weiguo Fan. Diamond in the Rough? Product Defect Detection and Summarization from UGCs
9:40am – 9:50am: Break
9:50am – 10:50am: Invited Talk by Yunyao Li
10:50am – 12:20pm: Panel
  • Behzad Golshan (Megagon Labs)
  • Thom Lake (Indeed.com)
  • Yunyao Li (IBM Research)
  • Vagelis Papalexakis (UC Riverside)
  • William Wang (UC Santa Barbara)
12:20pm-1:30pm: Lunch
1:30pm-2:30pm: Invited Talk by Vagelis Papalexakis
2:30pm-3:00pm: Session II – Oral Presentations (2 papers)
  • David Tuxworth, Dimosthenis Antypas, Luis Espinosa-Anke, Jose Camacho-Collados, Alun Preece and David Rogers. Deriving Disinformation Insights from Geolocalized Twitter Callouts
  • Yutong Li. Maps Search Misspelling Detection Leveraging Domain-Augmented Contextual Representations
3:00pm – 3:10pm: Break
3:10pm – 4:10pm: Invited Talk by William Wang
4:10pm – 4:50pm: Poster Session
4:50pm – 5:00pm: Closing Remarks

Call For Papers

We  encourage  submissions  that  describe  a  well-defined  piece  of  research  or  is thought-provoking. Topics will include but are not limited to information extraction, data cleaning, entity matching, schema matching, semantic search, summarization, language generation,  (common-sense) knowledge-bases and information seeking Q&A/Dialogue.

Submitted papers can be regular papers or extended abstracts. If there is sufficient interest from the authors of accepted papers,  we may publish the post-proceedings at CEUR. The maximum length of a regular paper is 8 pages plus unlimited number of pages for references. The maximum length of an extended abstract is 4 pages plus unlimited number of pages for references. At least one author of every accepted paper is expected to attend the workshop. Regular papers will be given an oral presentation slot.  Extended abstracts will be presented in the form of poster/demo/short talks, depending on the workshop schedule.


Important Dates

  • Submission deadline (All submission types): Jun 4th, 2021 (extended) May 20th, 2021
  • Notification of acceptance: June 10, 2021
  • Camera ready papers due: TBD



Papers should be formatted following KDD2021 template (as describe in guidelines here: https://www.acm.org/publications/proceedings-template) and submitted using the submission system available in this link: https://easychair.org/conferences/?conf=wit2021





If you have any questions or inquiries regarding the workshop or need further information, please do not hesitate to send an email to wit@megagon.ai.