S-LDA: Documents Classification Enrichment for Information Retrieval
Résumé
In recent years, the research on topic modeling techniques has become a hot topic among researchers thanks to their ability to classify and understand a large text corpora which has a beneficial effect on information retrieval performance, but recently user queries are more complicated because they need to know not only which documents are most helpful to them, but also which parts of documents are more or less related to their request. Also, they need to search by topic or document, not merely by keywords.
In this context, we propose a new approach of automated text classification based on LDA topic modeling algorithm and the rich semantic document structure which helps to semantically enrich the generated classes by indexing them in the documents sections according to their probabilities distribution and visualize them through a hyper-graph.
Experiments have been conducted to measure the effectiveness of our solution compared to topic modeling classification approaches based on text content only. The results show the superiority of our approach.