INFACT
The repository contains the source code for the INFACT evaluation framework, which is developed to conduct user-centric evaluation studies for dialog systems, Q&A, conversational recommender systems, and related domains.
Install / Use
/learn @ahtsham58/INFACTREADME
INFACT: An Online Human Evaluation Framework for Conversational Recommendation
This repository contains complete source code of the INFACT framework, which is developed to facilitate the user-centric evaluation studies of dialog systems, conversational recommender systems, and related domains.
Installation
Basically, To run the project on localhost, first you need to create and activate a virtual environment for your Django project. Therefore, please follow the below commands sequentially
cd CRS_Evaluation
python3 -m pip install --upgrade pip
pip3 install requirement.txt
pip3 install virtualenv
To verify the installation of virtual environment library, use
which virtualenv
Now, create a virtual environment for the INFACT framework using
virtualenv venv
Activate your venv
source ./venv/bin/activate
After this command, you should be able to see venv as a virtual environment created in your command promt.
Once you create and activate your virtual environment, to run the localhost, use
python3 manage.py runserv
Once your localhost server is ready, hit this URL in a browser and experience the wonders
http://127.0.0.1:8000/
How to parse the JSON data collected with this study?
The Python script named 'Parse_study_data.py' can be used to parse the complete data into MS excel sheets.
INPUT: JSON study data
OUTPUT: Separate excel sheets for valid, invalid rating scores, demographic, and feedback questionnaires.
Helpful links
You may have a look at the below links in case of abbration in activating the virtual environment
Install Django using virtualenv
Configure a virtual environment in PyCharm
Create a virtual environment in PyCharm terminal
Citation
@InProceedings{manzoor2022infact,
title={INFACT: An Online Human Evaluation Framework for Conversational Recommendation},
author={Manzoor, Ahtsham and Jannach, Dietmar},
booktitle={KaRS Workshop at RecSys '22},
address = {Seattle, USA},
year={2022},
keywords = {Conversational Recommender Systems, user-centric studies, evaluation, dialog systems},
}
Related Skills
node-connect
344.1kDiagnose OpenClaw node connection and pairing failures for Android, iOS, and macOS companion apps
frontend-design
96.8kCreate distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
openai-whisper-api
344.1kTranscribe audio via OpenAI Audio Transcriptions API (Whisper).
qqbot-media
344.1kQQBot 富媒体收发能力。使用 <qqmedia> 标签,系统根据文件扩展名自动识别类型(图片/语音/视频/文件)。
