Abstract:

In recent years, text mining has become more popular in the field of Education mostly because of the growing number of systems which store large data about students and their materials. Text mining technology provides a solution to bridge the knowledge gap between unstructured text and structured information representation.

Text mining uses techniques such as trends and associations to automatically retrieve, extract and discover statistics information in large data. It can help humans to identify and verify hidden information from text more efficiently and it can discover relationships in the huge data. A number of text mining tools have been developed to support research, yet none of them was designed for senior projects.

The volume and diversity of senior projects are expanding each year. It is exceptionally difficult for students and advisers to keep track of each project. They face difficulty to refer to the previous projects, extract some useful information and whether the proposed ideas match potential advisers and their research interests. In addition, decision-makers do not have valuable statistics about senior projects, assigned advisers, and areas of interest of senior projects. Therefore, there is a need to develop a tool that can archive and extract useful information about the archived projects automatically.

In this paper, we propose SaveIT that incorporates text mining techniques. SaveIT extracts valuable knowledge from titles and abstracts of the projects. Also, it visualizes some interesting information such as the most and least applied areas of interest among senior projects. In addition, SaveIT allows the user to extract the correlation among areas of interest and research trend analysis.

Our proposed solution is a web-based application that used for archiving a large number of senior project reports and prototypes of previous and upcoming years. It uses the power of text mining tools to generate statistics and interesting patterns from senior projects like trends of programming languages, the frequency plot of areas of interest, and the relationship among areas of interest represented in the network graph. SaveIT helps users without prior expertise in textual analysis to find answers to some important questions.

SaveIT targets students, advisers and researchers at the Information Technology department at King Abdulaziz University. All users will be able to get some knowledge about previous projects, projects areas, and the areas of interest of potential adviser. Results showed that the text mining tool was able to identify a considerable number of relevant terms from the texts analyzed, providing concise representations of projects which can support students and advisers.

Keywords:

Text-mining, senior projects, smart, archive.