INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
Text mining and Text analytics is a process of extracting use full information from Text Documents. Due to the extreme growth in online textual information.eg: “Email messages and social medias online News” To organizing this data is one of the major problem. so we organize and handle this type of E-learning documents we introduce a approaches called topic modeling and sentiment analysis these two methods are classify textual data and documents in a systematic manner, where Topic modeling implemented by the use of LDA(Latent Dirchlet Allocation) it converts the text documents into sentences and words then DTM(Document Term Matrix) assigns the frequency for each and every Term Present in Text document based on number of occurrences then LDA groups relevant Topics. Sentiment analysis implemented by using SVM classifier, SVM (Support vector Machine) mainly concentrate on opinion mining to find out the emotion of the particular person on that particular Document like happy, sad, satisfied, unsatisfied etc. Thus E-learning documents can be simply retrieved and classified using these methods which is also proven by experimental verifications. From the experimental result real world data from various areas shows that our proposed system out performs more than a few other baseline methods.
Keywords:
Topic modeling, LDA, classification, sentiment analysis, SVM, DTM, E-Learning, term frequency
Cite Article:
"Machine learning on text analytics and categorization through R-Language", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.3, Issue 3, page no.52 - 59, March-2018, Available :http://www.ijsdr.org/papers/IJSDR1803013.pdf
Downloads:
000337073
Publication Details:
Published Paper ID: IJSDR1803013
Registration ID:180074
Published In: Volume 3 Issue 3, March-2018
DOI (Digital Object Identifier):
Page No: 52 - 59
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn