2D text visualization for the retrieval of Malay documents

• Main Authors: Kamal Ismail, Normaly, Tengku Sembok, Tengku Mohd
• Other Authors: Jha, Manoj K.
• Format: Book Chapter
• Language: English
• Published: WSEAS 2012
• Subjects: QA76 Computer software
• Online Access: http://irep.iium.edu.my/28523/1/Normaly_Harvard_2012.pdf; http://irep.iium.edu.my/28523/; http://www.wseas.us/e-library/conferences/2012/CambridgeUSA/MATHCC/MATHCC-15.pdf

Search engine applications like Google and Yahoo present their results in the form of onedimensional linear list that usually comprise three times of the screen size per page and several number of pages. The results are displayed in the list of inconsistent declining ranks without displaying its rank values. The one-dimensional linear list display of the results data will cause classification of the results data meaningless. New queries relating to the original query are available, but its relationship strength values are not provided An application that can display all the result data in a two-dimensional text visualization within one page and circular form is proposed. The relationship strength of the result data with the query can be evaluated by finding the distance between the location of the result data to the center of the circle. Classifications that are made in the form of text and color can easily apply to the application. Malay translated Al-Quran and Malay translated hadith are used as corpuses for the application. Three functions in the application display the relationship between words and words, between words and documents, and between documents and documents. Various combinations of formulas can be used to find the values of these relationships that will be used as the rank values in the application. This, two-dimensional text visualization (TDTV), application is evaluated using two mechanisms. First, by solving a task and then, follow by answering the usability questionnaire. The results from the task section show that the variety of related documents can be retrieved in a reasonable time frame. The results from the usability questionnaire show about 75 percent of the respondents agree that the two-dimensional text visualization (TDTV) application is better than applications that display its results in one-dimensional linear list.