Portfolio Optimizer Tool (POT) Based on Trend Market Analysis and News Sentiment Analysis
Date
2014-10-16
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Stock market and portfolio analysis on financial data conducted by professional investors,
who seek informative decision making, is found to be complicated and hard to understand
by non-professional investors. In this thesis, we suggest to develop a portfolio optimizer
framework which is intended to be beneficial to investors ranging from professionals to non-
professionals. The proposed framework integrates data mining techniques such as frequent
pattern mining, clustering, classification, and sentiment analysis. To achieve the target, we
will apply sentimental analysis to investigate stock market trend in order to predict the
actual stock price index movement. This has been realized as an important task for the last
decades following the development in technology, though investments in the stock market
are active for over a century. According to the accuracy of prediction results, the investors
can be guided to investments that might help them make more money. On the other hand,
loss of investment is the risk when the prediction reported to investors suffers from lower
accuracy or when the prediction is missing some of the factors or objectives that directly
or indirectly affect the trend of the market. The integrated framework will help to find the
correlation between various stocks in the market to determine groups of stocks predicted
to behave similarly, i.e., maintain same trend whether up or down. We will concentrate on
the classification of stocks by sector and by industry. This will allow investors to diversify
their investments and hence reduce the risk if they prefer. In conclusion, this thesis provides
useful guidelines for investors to select appropriate market areas and formulate efficiently
diversified investment portfolios. The reported test results demonstrate the applicability and
effectiveness of the proposed framework.
Description
Keywords
Information Science
Citation
Al-jomai, R. (2014). Portfolio Optimizer Tool (POT) Based on Trend Market Analysis and News Sentiment Analysis (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/27388