Democritus University of Thrace (DUTH)
Τεχνικές ανάλυσης δεδομένων και αξιολόγηση διαδικτυακών εφαρμογών
Abstract
dc:descriptionThe rapid growth of Internet applications rendered imperative need their assessment. Factors of assessment constitute the degree and their way of usage from the users. The existence of indexes and metrics would facilitate the assessment of their usage. Moreover, the application of algorithms for the classification of users' preferences and the production of suitable proposals to the administrators, so that their content organisation to be optimised, would contribute positively to the evaluation of their usage. The aim of the present thesis is a global methodology for the usage assessment of Internet applications that will include: (a) the adoption of indexes and metrics for their usage analysis, (b) the production of algorithms for the usage classification and the generation of suggestions to the administrators, (c) the application of data mining techniques for data analysis of their usage and (d) the evaluation of the application of the methodology through (via) questionnaires to the users that will contribute to the confirmation of the analysis method insights. The existence of usage data constitutes a necessary prerequisite for the evaluation of each Internet application. The usage of e-learning data is an ideal case. Not only because of their easy availability, but also because they are provided in abundance and their continuous usage by the students. On the other hand, this data offer authenticity, because of their continuous recording in the log files. The main problem of Learning Management Systems (LMS) is that while they offer a wide range of work methods for the distribution of information and the communication of the participants in a course, the management and the exploitation of the contained information becomes difficult because of its volume. They are usually limited in certain reports that do not help however to the extraction (draw) of useful conclusions for the course or the participants in this, and they are useful only for administrative aims of each platform. Also, they do not offer concrete tools that would allow the assessment of users' actions and their training content. The particular methodology contributes to the existing knowledge and educational practice with the use of existing techniques in a different way. The new indexes and metrics, the algorithms for the classification of the applications usage and generation of suitable suggestions to the administrators, the application of data mining techniques for data analysis and the evaluation of the application of the methodology constitute the basic components of the contribution of the thesis. The proposed methodology of usage assessment and data analysis of Internet applications is proved effective. The use of classification and suggestions algorithms to the instructors helps them to proceed to suitable readjustments in the educational material of their courses. The application of data mining techniques contributes to become in-depth data analysis of the Internet application usage. The reliability of the methodology is confirmed with its evaluation. The fact that after the experiment both methods of evaluation converge to the same results, supports the proposed methodology and consequently assist the administrators of the content in the platform to advance in the application of concrete proposals for the improvement of the content both in quality and quantity. From pedagogic point of view, this methodology contributes to the improvement of courses content and their readjustment according to the students' capabilities. The improvement of courses content gives to the students the opportunity for asynchronous study with actual and optimal educational material. This methodology can constitute the basis for the design and the creation of an adaptive Internet application that would receive in real time the used data from it and will propose suitable suggestions of content readjustment and optimisation. Whether the field of the application of the methodology is limited only in the elearning platforms, it is needed to be investigated. The metrics enrichment, homogeneity and interest should also be used in other Internet applications, so that certain conclusions will be drawn about their possible application in them. The enrichment presents how much information is delivered to the final user, the homogeneity reflects the percentage of information that each user discovers independently and the interest shows if the users are pleased with the content of the web pages that visit and do not log out. The prerequisites for the application of the methodology are the content updates from the administrators and the existence of activity from the users of the Internet applications.
Degree
thesis:*- Grantor dc:publisher
- Democritus University of Thrace (DUTH)
- Year dc:date
- 2011
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Βαλσαμίδης, Σταύρος
- Valsamidis, Stavros
Subjects
dc:subject × 15- Διαδικτυακές εφαρμογές
- Δείκτες
- Μετρικές
- Αλγόριθμοι
- Τεχνικές εξόρυξης δεδομένων
- Αποτίμηση εφαρμογής μεθοδολογίας
- Internet applications
- Indexes
- Metrics
- Data mining techniques
- Evaluation of the methodology application
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/27792