Intelligent Stock Selecting via Bayesian Naive Classifiers on the Hybrid Use of Scientific and Humane Attributes

Among all kinds of investment activities, security´s transaction is an important activity among all investors´ involvements in the past decade. How to find out the relationships between a security´s name, price, trading quantity, and/or other scientific technical indices, humane feeling, and how these factors affect the buy or the sell timing is an important condition to be a successful investor. A Bayesian naive classifier is used to decide the future trends of a stock. 107 Data records were collected, a total of 9 attributes were used in this classification process. We exclusively take one thirds (30 examples) to test the validity of the develop inference model. The model shows a plain 57% of predicting accuracy and a high estimated possibility of 86.54% without losing his money by investing wrong targets. This result is helpful for those who have great interests to make profit in a stock market of similar situations.