<strong>Cooking is All About People: Comment Classification on Cookery Channels Using Bert and Classification Models (Malayalam-English Mix-Code)</strong>

Abstract

The scope of a lucrative career promoted by Google through its video distribution platform YouTube 1 has attracted a large number of users to become content creators. An important aspect of this line of work is the feedback received in the form of comments which show how well the content is being received by the audience. However, volume of comments coupled with spam and limited tools for comment classification makes it virtually impossible for a creator to go through each and every comment and gather constructive feedback. Automatic classification of comments is a challenge even for established classification models, since comments are often of variable lengths riddled with slang, symbols and abbreviations. This is a greater challenge where comments are multilingual as the messages are often rife with the respective vernacular. In this work, we have evaluated top-performing classification models and four different vectorizers, for classifying comments which are a mix of different combinations of English and Malayalam (only English, only Malayalam and Mix of English and Malayalam). The statistical analysis of results indicates that Multinomial Na&amp;iuml;ve Bayes, K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Random Forest and Decision Trees offer similar level of accuracy in comment classification. Further, we have also evaluated 3 multilingual sub-types of the novel NLP language model, BERT and compared its&amp;rsquo; performance to the conventional machine learning classification techniques. XLM was the top-performing BERT model with an accuracy of 67.31%. Random Forest with Term Frequency Vectorizer was the best the top-performing model out of all the traditional classification models with an accuracy of 63.59%.

Adapter-based fine-tuning of pre-trained multilingual language models for code-mixed and code-switched text classification

“…Most recently, fine-tuning pre-trained monolingual models such as BERT and multilingual models such as XLM-R and mBERT (Hande et al, 2021b) have been used. PLMs outperformed other Deep Learning and Machine Learning techniques in some studies (Chakravarthi et al, 2020;Aguilar et al, 2020;Khanuja et al, 2020), while they performed poorly in some others (Kazhuparambil and Kaushik, 2020). But according to Kazhuparambil and Kaushik (2020), PLMs can be made the top-performing models for CMCS data classification by optimizing hyper-parameters.…”

Classification of Code-Mixed Text Using Capsule Networks

“…The most notable one is the Spanish-English and Hindi-English code-mixed datasets released for the Se-mEval 2020 task (Javdan et al, 2020). Other than that, there are code-mixed datasets available between Malayalam-English (Kazhuparambil and Kaushik, 2020), and Tamil-English (Chakravarthi et al, 2020). Interestingly, other than the Spanish-English dataset, all the other datasets we identified involve Indic languages.…”

“…Interestingly, other than the Spanish-English dataset, all the other datasets we identified involve Indic languages. The Malayalam-English dataset was obtained from A food recipe dataset with about 3000 samples Kazhuparambil and Kaushik (2020). Figure 1 and the Table 1 show the class distribution and few samples of the Malayalam-English dataset, respectively.…”

“…it has some unlabelled data, however collecting labelled data is challenging). A food recipe dataset with about 3000 samples (Kazhuparambil and Kaushik, 2020) was used as the Malayalam-English code-mixed data. For Sinhala-English code-mixed data, a corpus of 10000 user comments in the domain of telecommunication was annotated with two types of information: aspects related to the telecommunication domain, and overall sentiment of the comment.…”

Money Often Costs Too Much: A Study to Investigate The Effect Of Twitter Sentiment On Bitcoin Price Fluctuation

“…There exist many challenges when performing sentiment analysis on the multilingual text as explained in [28], [29], [30]. The first challenge is to perform language identification.…”