A New Hybrid Technique on Image Emotion Classification System Using LBP Variants
Dr.Ramandeep Kaur
, Dr. Vijay Bharadwaj
TF-IDF, N-Grams, MSA, Feature Extraction, Sentiment Analysis, Gurmukhi script, Emotion Classification, Decision tree, Local binary pattern.
Abstract: Social media has become a common platform for exchanging opinions and ideas. Various Smartphone apps, such as Facebook, Twitter, Instagram, and WhatsApp, offer a common forum for people to communicate with one another. In this paper, LBP variants and a hybrid approach are considered for emotion classification on a Twitter image dataset containing images linked to the Gurmukhi script. This data was then filtered to only include content that fit into one of four emotion categories: happy, love, sad, or angry. For feature extraction, LBP and its variants such as MLBP, DRLBP, OCLBP, CSLBP, RLBP, and Hybrid of MLBP and DRLBP are used. The classification was done by using supervised classifiers such as decision tree, nave Bayes, kNN, and SVM, which classified the data into the appropriate emotion group. The resulting values were reported in order to ascertain the precision, recall, f-measure, and accuracy of the proposed method. The experimental results indicate that the decision tree is the highest, with high recall, F-score, and accuracy parameters. The Hybrid strategy is also measured to outperform the others
"A New Hybrid Technique on Image Emotion Classification System Using LBP Variants", IJNRD - INTERNATIONAL JOURNAL OF NOVEL RESEARCH AND DEVELOPMENT (www.IJNRD.org), ISSN:2456-4184, Vol.7, Issue 8, page no.1010-1022, August-2022, Available :https://ijnrd.org/papers/IJNRD2208109.pdf
Volume 7
Issue 8,
August-2022
Pages : 1010-1022
Paper Reg. ID: IJNRD_182497
Published Paper Id: IJNRD2208109
Downloads: 000118815
Research Area: Computer Science & Technology
Country: BATHINDA, Punjab, India
ISSN: 2456-4184 | IMPACT FACTOR: 8.76 Calculated By Google Scholar | ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 8.76 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: IJNRD (IJ Publication) Janvi Wave
