Comparison Of Multinomial Naive Bayes Algorithm And Logistic Regression For Intent Classification In Chatbot

Setyawan, Muhammad Yusril Helmi; Awangga, Rolly Maulana; Efendi, Safif Rafi

doi:10.1109/incae.2018.8579372

Cited by 41 publications

(20 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The minimum ACC was with 0.504 on hotel datasets; the minimum PRE was the tweets dataset with 0.62, the minimum F-MES and REC were on tweets with 0.556 and 0.504 Straight. Compared to the results in [15], Random Forest Classifier perhaps the most productive arrangement strategies when used in image, but it gives low results in text classification.…”

Section: G Random Forest Classifiermentioning

confidence: 90%

Performance Evaluation in Arabic Sentiment Analysis during the Covid-19 Pandemic

Sakr¹,

Amin

Grwany

2021

The Egyptian Journal of Language Engineering

View full text Add to dashboard Cite

This paper classifies sentiment analysis in Arabic language and mining sentiment in relation to the COVID-19 pandemic in the period (2019 -2021). Three large data sets are collected from tweets, hotel and restaurant reviews for building the proposed sentiment analysis model. We compared eight machine learning algorithms, Multinomial Naïve Bayes (MNB), Bernoulli Naïve Bayes (BNB), Decision Tree (DT), K-nearest neighbour classifier (KNN), Support Vector Machines (SVM), Linear Support Vector Classifier (LSVC), Random Forest Classifier (RFC) and Stochastic Gradient Descent Classifier (SGD) on three cases: n-gram unigram, bigram, and trigram for each algorithm. The performance evaluations are compared according to precision, recall, and Fmeasure.The polarity prediction results in sentiment analysis models were achieved by linear SVC using hotel dataset with bigram case, with the accuracy of 0.966, precision of 0.967, recall of 0.966 and F-measure of 0.966 . The rest algorithms give normal execution on all datasets. It may very well be reasoned that the AI calculations need the right morphological components to upgrade the classification exactness when managing various words that assume various parts in the sentence with a similar letter.

show abstract

Section: G Random Forest Classifiermentioning

confidence: 90%

Performance Evaluation in Arabic Sentiment Analysis during the Covid-19 Pandemic

Sakr¹,

Amin

Grwany

2021

The Egyptian Journal of Language Engineering

View full text Add to dashboard Cite

show abstract

“…They found that the SVM outperforms the rest of the machine-based approaches. The authors in [28] proposed classification on the chatbot application, used the NB method and compared it with the LR method to determine the class intention. The experiment results show the Logistic Regression model is a higher score than the Naïve Bayes model.…”

Section: Question Classification Algorithmsmentioning

confidence: 99%

Part-of-Speech tagging enhancement to natural language processing for Thai wh-question classification with deep learning

Chotirat¹,

Meesad²

2021

Heliyon

View full text Add to dashboard Cite

Question classification is a crucial task for answer selection. Question classification could help define the structure of question sentences generated by features extraction from a sentence, such as who, when, where, and how. In this paper, we proposed a methodology to improve question classification from texts by using feature selection and word embedding techniques. We conducted several experiments to evaluate the performance of the proposed methodology using two different datasets (TREC-6 dataset and Thai sentence dataset) with term frequency and combined term frequency-inverse document frequency including Unigram, Unigram+Bigram, and Unigram + Trigram as features. Machine learning models based on traditional and deep learning classifiers were used. The traditional classification models were Multinomial Naïve Bayes, Logistic Regression, and Support Vector Machine. The deep learning techniques were Bidirectional Long Short-Term Memory (BiLSTM), Convolutional Neural Networks (CNN), and Hybrid model, which combined CNN and BiLSTM model. The experiment results showed that our methodology based on Part-of-Speech (POS) tagging was the best to improve question classification accuracy. The classifying question categories achieved with average micro 𝐹 1 -score of 0.98 when applied SVM model on adding all POS tags in the TREC-6 dataset. The highest average micro 𝐹 1 -score achieved 0.8 when applied GloVe by using CNN model on adding focusing tags in the Thai sentences dataset.

show abstract

“…Dalam makalah ini, kami menggunakan Confusion Matriks untuk menguji hasil testing dengan memperhatikan akurasi, presisi, dan recall. Untuk menghitung akurasi pada confusion matriks menggunakan persamaan 3 [24]. Dari hasil pengujian algoritma Multinomial Naïve Bayes diketahui bahwa performance algoritma memiliki akurasi rata-rata sebesar 74%, precision sebesar 74% dan recall sebesar 74% dalam memprediksi sentimen terhadap penanganan Covid di Indonesia.…”

Section: Validasi Modelunclassified

Algoritma Multinomial Naïve Bayes Untuk Klasifikasi Sentimen Pemerintah Terhadap Penanganan Covid-19 Menggunakan Data Twitter

Yuyun

Hidayah

Sahibu

2021

RESTI

View full text Add to dashboard Cite

Currently, the spread of information Covid-19 is spreading rapidly. Not only through electronic media, but this information is also disseminated by user posts on social media. Due to the user text posted is varies greatly, it’s needs a special approach to classify these types of posts. This research aims to classify the public sentiment towards the handling of COVID-19. The data from this study were obtained from the social media application i.e., Twitter. This study uses a derivative of the Naïve Bayes algorithm, namely Multinomial Nave Bayes to optimize the classification results. Three class labels are used to classify public sentiment namely positive, negative, and neutral sentiments. The stage starts with text preprocessing; cleaning, case folding, tokenization, filtering and stemming. Then proceed with weighting using the TF-IDF approach. To evaluate the classification results, data is tested using confusion matrix by testing accuracy, precision, and recall. From the test results, it is found that the weighted average for precision, recall and accuracy is 74%. Research shows that the accuracy of the proposed method has fair classification levels.

show abstract

Comparison Of Multinomial Naive Bayes Algorithm And Logistic Regression For Intent Classification In Chatbot

Cited by 41 publications

References 11 publications

Performance Evaluation in Arabic Sentiment Analysis during the Covid-19 Pandemic

Performance Evaluation in Arabic Sentiment Analysis during the Covid-19 Pandemic

Part-of-Speech tagging enhancement to natural language processing for Thai wh-question classification with deep learning

Algoritma Multinomial Naïve Bayes Untuk Klasifikasi Sentimen Pemerintah Terhadap Penanganan Covid-19 Menggunakan Data Twitter

Contact Info

Product

Resources

About