Robust Machine Learning Techniques For Document Summarization

Computer Engineering

Currently huge amount of data is available on the internet which is increasing exponentially day by day. It becomes time consuming and tedious job to search a specific topic from the heap of information available. Document summarization is the key solution to the above stated problem. It refers to reducing the size of the document still preserving the main information of it. . Now to summarize the data using the computer program or algorithm is called the automatic document summarization. Abstractive and Extractive are the two main automatic document summarization techniques. To summarize the data mainly there are two steps, pre-process the data and process it, where in pre-processing the data is cleaned removing unwanted data and in processing various techniques are applied to summarize the data. Data summarization has various real life applications and is very useful for everyday life. This paper gives the hybrid approach for Multi-document Summarization where initially the documents are clustered using the effective grouping through advance similarity measure which also considers the dissimilarity of each document with every other document in the corpus, then the extraction technique is used for sentence extraction as they are reordered according to their weights obtained. Lastly the summary is effectively generated for each cluster.

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