Construction and Verification of Arts Data Preprocessing Framework For E Learning

dc.contributor.guideKumaravel, A
dc.coverage.spatial
dc.creator.researcherSasiKumar, C S
dc.date.accessioned2022-01-05T05:01:54Z
dc.date.available2022-01-05T05:01:54Z
dc.date.awarded
dc.date.completed2021
dc.date.registered
dc.description.abstractResearch studies on data preprocessing is crucial as it determines and highly influences the quality of prediction in the later phases. Hence the proposal for recommending a unique and practical combination in the spectrum of data preprocessing task types framework and it contains the following four technical aspects briefly justified. Firstly the A-aspect denotes the attribute selection based on modern research which makes comprehensive efforts on attribute selection processes to achieve effective preprocessing by reducing the data search space considerably. The attribute selection extends out both vertically and horizontally, because of increasing demands for dimensionality reduction and regulate on space complexity. The exploration of space as been reduced very well by trimming the insignificant attributes. Secondly in the R-aspect randomized subset of big datasets are applied for the purpose namely the improved efficiency. Different approaches for tackling the same problem yields more powerful validation of the generated model. Thirdly in the T-aspect the input data format is not as structured as the previous aspects, but presented as a collection of text files on student feedbacks and hence text mining techniques are applied along the rough set theory. And finally the S-aspect is meant for synthetic datasets in order to have better accuracy of models. Applying this framework on the datasets in the domain of e-learning renders appreciable amount of increment in the accuracy of classifiers which will be beneficial for e-learning industry. Motivated by the need for a framework which helps to perform the pre-processing trying in different ways to improve the accuracy of models, as primary objective of this work, are generated by Machine learning algorithms further on decision making in eLearning environment. As a result, its obtained better accuracy comparing to previous models tried for the same purpose in different approaches. Keywords: - Selection of Attributes, Search methods Text Mining, n-grams,; Syntheti
dc.description.note
dc.format.accompanyingmaterialDVD
dc.format.dimensions
dc.format.extent
dc.identifier.urihttp://hdl.handle.net/10603/354282
dc.languageEnglish
dc.publisher.institutionDepartment of Engineering and Technology(Computer Science and Engineering)
dc.publisher.placeChennai
dc.publisher.universityBharath University
dc.relation
dc.rightsuniversity
dc.source.universityUniversity
dc.subject.keywordComputer Science
dc.subject.keywordComputer Science Artificial Intelligence
dc.subject.keywordEngineering and Technology
dc.titleConstruction and Verification of Arts Data Preprocessing Framework For E Learning
dc.title.alternative
dc.type.degreePh.D.

Files

Original bundle

Now showing 1 - 5 of 13
Loading...
Thumbnail Image
Name:
80_recommendation.pdf
Size:
650.11 KB
Format:
Adobe Portable Document Format
Description:
Attached File
Loading...
Thumbnail Image
Name:
cert.pdf
Size:
233.62 KB
Format:
Adobe Portable Document Format
Loading...
Thumbnail Image
Name:
chapter 1.pdf
Size:
653.04 KB
Format:
Adobe Portable Document Format
Loading...
Thumbnail Image
Name:
chapter 2.pdf
Size:
297.12 KB
Format:
Adobe Portable Document Format
Loading...
Thumbnail Image
Name:
chapter 3.pdf
Size:
3.95 MB
Format:
Adobe Portable Document Format

License bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.79 KB
Format:
Plain Text
Description: