• SUJATA SUVARNAPATHAKI

Nifty 50 stocks are a group of top-performing equity stocks on the National Stock Exchange of India. It’s a good idea for investors to invest their capital in components of Nifty 50. It may reduce the investment risk to some extent since the Nifty 50 involves a varied range of stocks. Investors are often interested in understanding the behaviour of stocks, in general. A prior idea of the worth of a stock or the direction of the movement of stock would always benefit the investors. The stock market analysis enables the investors to make the informed decision of “Invest”, “Do not invest” or “Hold (If already invested)”. For this purpose, the investors may rely on significant fundamental analysis parameters and technical analysis parameters for understanding the behaviour of the stocks based on the historic data. But the public sentiments may also play a big role in behavioural finance and the financial decisions of investors may be influenced by the public sentiments. The economic indicators and the public sentiments are generally interrelated. With access to social media like Facebook, Twitter etc. people voice their opinions publicly about social, and political events and express their feelings. These data on social media are unstructured, hence may not be analyzed statistically but still may be useful. This research paper includes the use of unstructured and structured data for the segmentation of Nifty 50 stocks. The structured data comprises one fundamental analysis parameter, viz. ‘Price to Earnings Ratio’ (P/E Ratio) and one technical analysis parameter viz. ‘Relative Strength Index (RSI), for the segmentation of Nifty 50 stocks. Sentiment analysis is used as a Text Mining technique to extract information from the unstructured data from Twitter. The unstructured data in the form of tweets, for each stock in Nifty 50 is used for deriving an additional parameter Sentiment Score using which the ‘Proportion of Positive sentiment’ is computed. The data on 100 recent tweets about the stocks in Nifty 50 is fetched. The segmentation of Nifty 50 stocks into k clusters is then achieved using the k-means clustering method, based on the three variables, viz. RSI, P/E Ratio and the Proportion of Positive Sentiment. These “k” clusters of Nifty 50 stocks can help the investors to decide on trading actions like “Invest” or “Do not invest” or “Hold (If already invested)”. The above analysis is carried out using R programming.

Unstructured data, Text Mining, Sentiment Analysis, Twitter, Stock market, Cluster analysis, R

"USING UNSTRUCTURED DATA WITH STRUCTURED DATA FOR SEGMENTATION OF NIFTY 50 STOCKS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 6, page no.d184-d188, June-2022, Available :http://www.jetir.org/papers/JETIR2206323.pdf

"USING UNSTRUCTURED DATA WITH STRUCTURED DATA FOR SEGMENTATION OF NIFTY 50 STOCKS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 6, page no. ppd184-d188, June-2022, Available at : http://www.jetir.org/papers/JETIR2206323.pdf