An optimized topic modeling question answering system for web-based questions

The ability of the system to answer the searched formal queries has become active research in recent times. However, for the wide range of data, the answer retrieval process has become complicated, which results from the irrelevant answers to the questions. Hence, the main objective of the current article is a Topic modelling Question Answering (QA) system for the web-based Job posting data. Therefore, a novel Dove-based Alex net Gibbs Modeling (DbAGM) has been implemented for accurate answer retrieval and topic modelling for web-based questions in job assistance. The initialized job post data are filtered to remove the noise, and question features are extracted at the feature analysis phase. Furthermore, in the QA module, the extracted features are matched with the trained features set to retrieve the relevant answers for the searched web-based Question. Furthermore, the Topic is selected for the displayed answer keywords. The designed system is executed in the Python platform, and performance metrics are calculated. The proposed Model gained an accuracy rate of 95.6%, which is quite better than traditional models. It proved the suitability of the implemented Model for the specific Topic modelling QA system.