Visual Stock Information System Based On Ontology

With the development of internet and economics, there are more and more different types of stock information in the internet. They are very complex for high dimension, large volume and more changeable, which confuses the general users. The users need a friendly interface to the web and the computer.Ontology and visualization technology are used to solve the heterogeneous problems of stock information followed by development of Visual Stock Information System (VSIS). Visualization is the technique of producing a graphics, image for conveying the message to the user and capture human interaction for data analysis in the Stock Information System (SIS). Visualization technology used to provide easy accessible profile for stock data and stock information. Ontology based involves the use of ontology to combine stock data with stock information. The study provided a comprehensive ontology-driven of stock information, building semantic applications and specifying ontology domain of information.The study object focused on Yahoo Finance website (http://www.yahoo.com/finance). The website provides enough information that meets the goals and objectives of this study. Stock transaction data and stock information, were the two main kind of information focused. The stock information provided by yahoo is based on the text based model and tabular model, with the only one graphical mode shown from the chosen company. The stock information are scattered everywhere in the page, which is not easy for user to capture the acquired information. Only selling price and stock volume occupying the graphical model were shown in the website. Where by others stock historical data such as low price, high price, opened price and closed price historical data are presented in the tabular mode. Others stock information is all based on text base model and tabular model. The yahoo website does not allow user to visualize their stock services that seems to be difficult for user to understand and analyze other stock information based on the historical data, of the mentioned stock data and current information. Ontology modeling technology can conquer the heterogeneous information of the stock data. This is due to the fact that, ontology and semantic technology are increasingly used to represent intended ontology modeling of data and semantic services. It provides a means to overcome the stock heterogeneous by producing the explicit and formal description of stock domain and their relationship that exist in the domain of concept and the shared vocabulary among the domain members. The contributions of this study were discussed as follow:(1) VSIS requirements were given in detail. They included semantic ER Model, database design, the top use case diagram and the use case summary table in VSIS.(2) Ontology learning in VSIS:The VSIS ontology learning facilitates the construction of ontologies by the ontology engineering. The stock information extracted from the yahoo website was filtered, interpreted and verified in the protege which is open source tool developed by Stanford university. It is a tool for ontology representation using Web ontology Language (OWL) and Resource Description Framework (RDF) and RDF schema. The learning ontology provided VSIS structure that supported Ontology model with a set of ontologies in the Ontology Pools and Domain of knowledge.(3) Ontology definition in VSIS:Ontology is a formal representation of knowledge as a set of concepts within a domain, and the relationships between those concepts. In other words, ontology is described as a formal explicit description of concepts in a domain. Properties of each concept describes various features and attributes of the concept and restrictions on slots--Ontology+set of individual instances=knowledge base. Formally ontology was defined as a tuple, O=(C, R, A, I). The VSIS ontology used to describe the stock elements (instance, term and vocabulary), relationships, rules that build the VSIS domain. VSIS ontologies were equated with taxonomic hierarchies of classes, which defined the hierarchical relation from the root level to the node of instances in the stock domain.(4) VSIS framework based on ontology VSIS is a client server firmware that can be installed in the server and serve to several clients across the network users. VSIS provided a user interface that permitted user to interact with it, which was organized according to the user's requirements under client-server network environment. (5) VSIS prototype system VSIS prototype was developed under Windows operating system, Matlab version 7, and SQL Sever, which was able to search, retrieve and browse the stock information with different types.The study was able to examine all necessary measures for developing VSIS based on ontology model and user's perspective. And the emphasis was to determine whether an interactive model of user interface can have an impact on the user's perception on the stock data on the ontology environment. The results showed that VSIS has great ability to support user interaction. Based on the nature of information, VSIS provided with four layers of stock visualization, such as transaction data layer, non-transaction data layer, stock information and intelligent information layer. Different techniques of stock visualization were given, such as normal plotting methods, bar graph, plot graph polar and area according to the nature of stock information. Moreover in the transaction system of VSIS model user were allowed to define and select the interested model of visualization, compare transaction's profit and loss between the companies. Non-transaction data, intelligent information and other stock information, other graphical and tabular mode of visualization were used to present the required information. The study was able to demonstrate the prototype application to embody the requirements design of VSIS. The results showed that VSIS could provide the visual semantic and ontology based solution for stock data structure.