Artifical neural networks approach to short text clustering

Abstract

With the rapid growth of technology within the last few years, the need of dealing with data are also rising. Data exists everywhere in our life in different kind of forms. One of the most common type of data is text files. Having a deep look on datasets help us to understand those data and collect a lot of useful information about it. We are looking for a way to deal with short text data using neural network as a helping tool. The goal of this thesis is to propose an approach to cluster unlabeled data into the sets of data using some similarity measures. This thesis uses various number of methods to achieve it’s goal like neural networks, TF-IDF, SOM, K-Means,DBScan, ect. We will try to compare them to see the advantages and disadvantages of those method on clustering the same dataset.