A Comprehensive Comparison of Machine Learning Based Methods Used in Bengali Question Classification

Anika, Afra; Rahman, Md. Moklesur; Islam, Salekul; Jameel, Abu Shafin Mohammad Mahdee; Rahman, Chowdhury Rafeed

doi:10.48550/arxiv.1911.03059

Cited by 1 publication

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar implications are applicable for the other five models. Our method shows superior performance when it comes to finer class classification compared to the finer class classification results provided in [1] (Results provided in Table 4). This has been possible, because each of our six SGD models has the advantage of specializing on the finer classes of only one coarse class.…”

Section: Resultsmentioning

confidence: 85%

“…It is to note that we have not eliminated stop words. In [1], all the machine learning based algorithms performed better when stop words were not eliminated. Table 2 shows the average result of precision, recall and f1-score after 10 fold cross-validation for coarse class classification.…”

Section: Resultsmentioning

confidence: 99%

“…Lexical features and syntactical features were used to classify questions into appropriate classes [23]. In [1], the authors provided a comparison of machine learning-based methods based on performance and computational complexity. They used 7 different classifiers to conduct comparison where Stochastic Gradient Descent (SGD) performed the best.…”

Section: Question Classification Systems In Bengali Languagementioning

confidence: 99%

“…We keep stop words of question samples while classifying. In [1], the authors showed superior performance of different classifiers when stop words were not removed.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Hybrid Approach Towards Two Stage Bengali Question Classification Utilizing Smart Data Balancing Technique

Rahman,

Amin

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Question Classification (QC) system classifies the questions in particular classes so that Question Answering (QA) System can provide correct answers for the questions. We present a two stage QC system for Bengali. One dimensional convolutional neural network (CNN) based model has been constructed for classifying questions into coarse classes in the first stage which uses word2vec feature representation of each word. A smart data balancing technique has been implemented in this stage which is a plus for any training dependent classification model. For each coarse class classified in the first stage, a separate Stochastic Gradient Descent (SGD) based classifier has been used in order to differentiate among the finer classes within that coarse class in stage two. TF-IDF representation of each word has been used as feature for each SGD classifier separately. Experiments show the effectiveness of this two stage classification method for Bengali question classification.

show abstract

Section: Resultsmentioning

confidence: 85%

Section: Resultsmentioning

confidence: 99%

Section: Question Classification Systems In Bengali Languagementioning

confidence: 99%

“…We keep stop words of question samples while classifying. In [1], the authors showed superior performance of different classifiers when stop words were not removed.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Hybrid Approach Towards Two Stage Bengali Question Classification Utilizing Smart Data Balancing Technique

Rahman,

Amin

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

A Comprehensive Comparison of Machine Learning Based Methods Used in Bengali Question Classification

Cited by 1 publication

References 0 publications

A Hybrid Approach Towards Two Stage Bengali Question Classification Utilizing Smart Data Balancing Technique

A Hybrid Approach Towards Two Stage Bengali Question Classification Utilizing Smart Data Balancing Technique

Contact Info

Product

Resources

About