InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

A System and Method for Recommending Improvements to Questions on Knowledge Sharing Forums.

IP.com Disclosure Number: IPCOM000250454D
Publication Date: 2017-Jul-19
Document File: 4 page(s) / 148K

Publishing Venue

The IP.com Prior Art Database

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 41% of the total text.

A System and Method for Recommending Improvements to Questions on Knowledge Sharing Forums

Abstract for the article

Disclosed is a system which can, as the user is formulating or has formulated questions, automatically recommends edits to the question on the knowledge sharing forums. These recommendations are based on history of edits on the similar scenarios as performed by reputed users and via comments asking for clarifications and more info in the question.

Knowledge sharing (KS) question answering forums are the mechanism to exchange knowledge (information, skills, or expertise) around the globe between people, communities and organizations. The KS forums aim to build a reusable knowledge base and to achieve this, well- formed questions are indispensable. Thus, each KS forum has distinct & strict guidelines for the acceptable format of the questions and these are stringently followed by its community. As the count of newbies on KS forums is in preponderance and who are not familiar with these guidelines, they end up posting a large number of questions which either get disliked, closed voted, or are left unanswered.

Closed, Unanswered and Disliked (CUD) questions hurt both newbies and KS forums

 Newbies get disappointed as they are deprived of the answers which results in the bad first impression of the platform

 KS forums’ content quality degrades with the increasing number of CUD questions

The KS forums’ users with CUD questions can get answers only by properly editing their questions. In such scenarios, the only way out is to educate the users, the art of editing their CUD questions. This necessitate a system which can assist users by recommending the suitable edits for their CUD questions.

Presently, there is no such system that focus upon improvements of CUD questions on KS forums. In the proposed system, the edits are recommended to the CUD questions, in terms of some edit features, by learning the phraseology and mode of expression of questions based on three key facets:

1. Edit features (Fedit) based on o Frequently used text features for e.g.,

 orthographic – pronouns, articles, adverbs, prepositions etc.  formatting – special symbols, punctuations etc.

o Terminology & formatting guidelines imposed or practiced by the forum for e.g.,  Give live examples using multimedia, URLs or code snippets  Use popular tags, short and precise title  Avoid forbidden words, new or defunct tags, question words in title

2. Edit recommendations based on experts or reputed users of the forum o The users build a reputation based on

 Quality of their questions  Usefulness of their answers

o The reputation of users reflects the trust of the community in them o Most of the forums have a direct measure for the reputation for e.g.,

 “Reputation Points” (Stack Exchange)  “Level” (Yahoo Answers),  “Experts” (AnswerHub, Senexx SolvePath)

o The reputation can also be derived using the counts of  UpVotes, DownVotes, Likes, Dislikes, Views, Questio...