Design and Development of Topic Identification Using Latent Dirichlet Allocation

Data storing and retrieving are the most important task in the present condition. Storing can be ended based on the topic that the document describes. Text mining generates documents from the collection of topics. To identify the topics, we have to categorize the documents; to classify, we are using topic modeling. Text mining technique is used for discovering latent semantic structure which is a fragment of topic modeling. Various research areas that make use of probabilistic modeling includes software engineering, political science, and medical science. A topic model is a probability-based model that discovers the major themes which are a group of documents. The main idea is to treat the documents as mixtures of topics in the topic model, and every topic is viewed as a probability distribution of the words. This research work aims to propose a model called topic modeling using LDA, and this model has been experimented on two datasets, where one is two news group dataset, and other is twenty news group dataset, and finally, all the results are tabulated.