Böcker av Raunak Rathi

Text mining has emerged as one of the most important data processing activities over the last few decades. While it makes the life of millions of everyday users of digital plat- forms and applications much easier, it is a domain that also challenges researchers in numerous ways. The challenges are many fold - ranging from the volume of the data that needs to be processed, storage issues, language identi¿cation challenges and many more. The work in this thesis focuses on one particular aspect of text mining e.g. Key- word identi¿cation for a document. While this may seem to be quite a trivial activity for short passages, it is quite dif¿cult to successfully identify keywords for extremely long text documents. Doing so using an automated systems only adds to the challenge. Everyone in today's world understands the importance of data. In the context of business, data is used to analyze market trends or can be used to understand customer needs. It also helps to understand the user's perspectives and choices. There are var-ious ways that data plays a crucial role in our everyday lives. Most businesses would be bound to fail if they could not comprehend the data that was available. This data could vary from stock indices, to customer feedback, to worker sentiments and numerous other insights. Analyzing data also helps in advertisement noti¿cations or to suggest a piece of relevant information to the user. It also helps to understand the likes and dislikes of a user. It can make for a world with a better user experience in terms of an individuals needs, e.g., if a user is more interested in cricket, we can provide targeted insight to the user about cricket. A customized user experience for a user is more at- tractive than a bland user experience which is homogeneous for everyone. Everyone's needs are different from others as everyone has different perspectives and opinions. We offered examples of keyword extraction, the challenges involved and the major issues faced by designers of keyword extraction algorithms. Finally about some common application areas where keyword extraction is being used in real life scenarios. Attracting users and providing them with better services through relevant data also helps the system to understand the users' needs. A user consciously or unknowingly provides his information for use in business or expresses his views on various platforms. If a user expresses some political opinions, it helps us to tailor his experience better the next time he uses the system.