Online panel data quality: a sentiment analysis based on a deep learning approach

Youb Ibtissam, Azmani Abdallah, Hamlich Mohamed


The rise of online access panels has profoundly changed the market research landscape. Often presented by their owners as very powerful tools, they nevertheless raise important scientific questions, particularly regarding the representativeness of the samples they produce and, consequently, the validity of the information they provide. In this paper, we present an innovative approach, based on deep learning and sentiment analysis techniques, to assess in real time the representativeness of an online panel sample. The idea is to measure the extent to which the opinions of an online panel converge with opinions on social networks. To validate the proposed method, we conducted a case study on the emerging discussion on coronavirus disease (COVID-19) vaccination. The results not only proved the representativeness of online panel sample, but also demonstrated the feasibility and effectiveness of our approach.


Deep learning; Online panel; Representativeness; Sentiment analysis

Full Text:




  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

IAES International Journal of Artificial Intelligence (IJ-AI)
ISSN/e-ISSN 2089-4872/2252-8938 
This journal is published by the Institute of Advanced Engineering and Science (IAES) in collaboration with Intelektual Pustaka Media Utama (IPMU).

View IJAI Stats