A Prototype Of Web Application For Applicants To Match Graduate Schools Based On Data Mining

Recently, most college seniors look for graduate education as either their firstchoice or a backup plan. But it is hard to find a service for the applicants to findgraduate schools. Because of this, manually searching the graduate schools is theonly option for these applicants. It is technically inefficient and time consuming tofind all the possible graduate schools on the website. Hence, a web applicationneeds to be built to serve the immediate need of college seniors, help them retrievethe matching schools and narrow down their search efficiently.The web application utilizes data mining technology that matches the graduateschools according to the inputs of their career interest and academic preparation.After that it will display the results to the final users in three categories. Thetraining data set in this web application is acquired from the historical informationof the student who have been matriculated successfully by a particular graduateschool. Then use analysis tool Weka to analyze these data. Three data miningalgorithms–decision tree, na ve Bayesian and nearest neighborhood–have beenpicked out to build the model. After comparing them in both subjective andobjective ways, decision tree algorithm is the optimal option. C4.5, one of decisiontree algorithms, is used to build the model. Last but not least, the web application isimplemented in using that data mining engine.The web application utilizes data mining technology to build the retrievingmodel, uses the web tool WordPress to construct the whole website structure,employs PHP, JS technology to achieve the logical communication, applys Apacheas the server and MySQL as the database.