Task8: Semantic Question Similarity in Arabic

Task Description:

The goal of this task is to predict whether two questions are similar or not.

Input: A question pair.

Output: yes, if the two questions are similar or no otherwise.

Examples:

question1question2 semantically similar?
ما معنى الهجرة السرية؟ ما هي الهجرة السرية؟ yes
ما هي العوامل التي تؤثر في انتماء الشخص للوطن؟ كيف ينمو شعور عندم الانتماء للوطن؟ yes
ما طريقة تحضير كيك بالفواكه ؟ كيف احضر كيك بالفواكه؟ yes
ما هي عاصمة أوكرانيا؟ ما اسم عاصمة سويسرا؟ no
بم يتميز تطبيق الفايبر؟كيف أعمل تسجيل خروج من الفايبر؟ no
ما معنى الهجرة ؟ ما هي الهجرة ؟ yes
إلى ماذا يعود سبب تسمية الأردن؟ ماذا تسمى الأردن رسمياً؟ no

Data and tools:

Training data: https://ai.mawdoo3.com/nsurl-2019-task8

The task evaluation baseline(s) : A baseline will be developed using basic traditional statistical learning methods.

The evaluation procedure : A labelled holdout set will be hidden from participants. F1-Score will be used to measure the performance for each model.

Important Dates: http://nsurl.org/importantdates/

Task participation:

To participate in this task , the team leader has to do the following:

  1. Choose a name for your team (The name should reflect your team)
  2. login as an author to https://easychair.org/conferences/?conf=nsurl2019
  3. add the paper title as : <Team-name> at NSURL-2019 Task 8: Semantic Question Similarity in Arabic
  4. Paper authors of the paper: The team members
  5. Paper abstract and keywords: add a simple tentative abstract that you can modify anytime
  6. submit

Results:

After Evaluation

Paper submission:

After Evaluation

Task Organizers:

If you have any queries regarding this task, please refer to the task organizers:

Hussein Natsheh: <h.natsheh@mawdoo3.com>
Haitham Seelawi: <haitham.selawi@mawdoo3.com>
Ahmad Mustafa: <ahmad.mustafa@mawdoo3.com>
Wael Farhan: <wael.farhan@mawdoo3.com>