Publication:
User traversal behaviour mining of server logs using fuzzy FRS

dc.contributor.affiliation#PLACEHOLDER_PARENT_METADATA_VALUE#en_US
dc.contributor.authorRosli bin Omaren_US
dc.date.accessioned2024-10-08T07:37:06Z
dc.date.available2024-10-08T07:37:06Z
dc.date.issued2018
dc.description.abstractWeb Usage Mining (WUM) is the application of data mining methods in extracting potentially useful information from web usage data. Its application includes improving website design, personalised service, target marketing etc. Even though there has been an extensive study in WUM, lack of related product commercialisation indicates that there are still a number of outstanding research issues in this area. Among the challenges mentioned in the literature include inefficiency in mining typically large weblogs, extracted patterns that are not representative of actual user behavior, and mining results which are too general, uninteresting and lack insights. This thesis attempts to address the above problems in three parts. Firstly, based on the notion of regularity, a mining algorithm is introduced to efficiently extract usage patterns from large weblogs that are reflective of individual user behaviour. Secondly, a fuzzy method is incorporated into the algorithm that enables the expression of the pattern quality, thus reducing possible confusion due to extremely large number of patterns. Finally, in order to gain deeper insights of the extracted patterns, the algorithm is further extended using the framework of transitional pattern to capture possible variation in pattern behaviour, thus facilitating the subsequent pattern interpretation process. The promising results obtained from a series of experiments conducted suggest that the new algorithm is faster and more scalable compared to an existing one, especially when mining large weblogs. Furthermore, the extracted patterns demonstrate better representation of user traversal behaviour, contain less ambiguity, and are more readily interpretable for subsequent analysis.en_US
dc.description.callnumbert QA 76.9 D343 R788U 2018en_US
dc.description.degreelevelDoctoralen_US
dc.description.identifierThesis : User traversal behavior mining of server logs using fuzzy FRS /by Rosli bin Omaren_US
dc.description.identityt11100396745RosliOmaren_US
dc.description.kulliyahKulliyyah of Information and Communication Technologyen_US
dc.description.notesThesis (Ph.D)--International Islamic University Malaysia, 2018.en_US
dc.description.physicaldescriptionxiv, 168 leaves :illustrations ;30cm.en_US
dc.description.programmeDoctor of Philosophy in Information Technologyen_US
dc.identifier.urihttps://studentrepo.iium.edu.my/handle/123456789/9296
dc.identifier.urlhttps://lib.iium.edu.my/mom/services/mom/document/getFile/snSDOgxs10SVxEpTAMCuvWqHGmBqYB8v20190315115531364
dc.language.isoenen_US
dc.publisherKuala Lumpur :International Islamic University Malaysia,2018en_US
dc.rightsCopyright International Islamic University Malaysia
dc.subject.lcshData miningen_US
dc.subject.lcshWeb usage miningen_US
dc.subject.lcshFuzzy algorithmsen_US
dc.titleUser traversal behaviour mining of server logs using fuzzy FRSen_US
dc.typeDoctoral Thesisen_US
dspace.entity.typePublication

Files

Original bundle

Now showing 1 - 2 of 2
Loading...
Thumbnail Image
Name:
t11100396745RosliOmar_SEC_24.pdf
Size:
320.43 KB
Format:
Adobe Portable Document Format
Description:
24 pages file
Loading...
Thumbnail Image
Name:
t11100396745RosliOmar_SEC.pdf
Size:
1.5 MB
Format:
Adobe Portable Document Format
Description:
Full text secured file

Collections