E. mokymosi kursų duomenų analizė, taikant duomenų gavybos metodus
LTSparčiai didėjant apdorojamų ir saugomų duomenų kiekiui, aktuali tampa duomenų analizė, padedanti priimti greitus, efektyvius ir teisingus sprendimus. Duomenų analizei atlikti taikomi duomenų gavybos metodai. Duomenų gavybos metodais iš duomenų išgaunamos naudingos žinios. Straipsnyje apžvelgiamos duomenų gavybos technologijų taikymo e. mokymesi galimybės, atrenkami e. mokymosi kurso duomenys ir parenkami duomenų gavybos metodai e. mokymosi kurso duomenų analizei atlikti. Pristatomi atlikto tyrimo, kuriame duomenų gavybos metodais analizuojami e. mokymosi kurso duomenys, rezultatai. Atlikto tyrimo rezultatų pagrindu apibrėžiamos e. mokymosi kurso tobulinimo kryptys.
ENData mining means searching for certain patterns within large sets of data, which creates a lot of possibilities for decision makers. By analysing those patterns, better decisions can be made in order to improve e-learning process. The research interest in using data mining in e-learning is constantly increasing. According to L. Shen, M. Wang, R. Shen the database of e-learning system includes much useful information which can be effectively used for the improvement of e-learning process. Authors emphasise that due to the vast quantities of data these systems can generate daily, it is very difficult to analyse this data manually and a very promising approach towards this analysis objective is the use of data mining techniques. The purpose of this research is to analyze the course data using data mining techniques in order to provide course development trends to more efficient and higher quality studies. Research methods: the analysis of scientific literature, data analysis using data mining techniques. In this research, the data of Database design course were analysed. All data necessary for this research were transferred from Moodle database. There were created the procedures to perform calculations. Selected data set consists of 11 fields. The sources of selected data are analysable course records: the number of performed self-test, time spent for self-test, average of test grades, average of control works grades, final mark and so on. WEKA tool was used to perform data analysis using data mining techniques. For data analysis was chosen descriptive data mining model‟s techniques: clustering and association rules. For the clustering in this work was used k means algorithm, for the finding association rules was used Apriori algorithm.After clustering technique were obtained statistical data which showed that the students with excellent, very good or good final grade actively studied theoretical training materials, performed self-tests, better performed control tests, control works, exam tasks. Using the association technique was obtained that the students who get minimal (the worst) grades did the course out of continuity; they gave little attention to the study of theoretic materials, self-tests; they passed the final exam without gaining necessary theoretic knowledge. The students who got good marks successfully completed all sections of theoretic materials passed self-tests and passed the final exam only on completing all course activities. Data analysis showed the disadvantages of course structure. It was resolved that Database design course should be modified introducing the learning process control.
