295 Hits in 3.1 sec

Experimental Evaluation of Deep Learning models for Marathi Text Classification [article]

Atharva Kulkarni, Meet Mandhane, Manali Likhitkar, Gayatri Kshirsagar, Jayashree Jagdale, Raviraj Joshi
2021 arXiv   pre-print
This works aims to provide a comprehensive overview of available resources and models for Marathi text classification.  ...  We evaluate CNN, LSTM, ULMFiT, and BERT based models on two publicly available Marathi text classification datasets and present a comparative analysis.  ...  We would like to express our gratitude towards our mentors at L3Cube for their continuous support and encouragement.  ... 
arXiv:2101.04899v2 fatcat:g6sdkrfvcrbexasz6kbawvnphq

An Attention Ensemble Approach for Efficient Text Classification of Indian Languages [article]

Atharva Kulkarni, Amey Hengle, Rutuja Udyawar
2021 arXiv   pre-print
Experimental results show that the proposed model outperforms various baseline machine learning and deep learning models in the given task, giving the best validation accuracy of 89.57\% and f1-score of  ...  This paper proffers team SPPU\_AKAH's solution for the TechDOfication 2020 subtask-1f: which focuses on the coarse-grained technical domain identification of short text documents in Marathi, a Devanagari  ...  Such results, however, for Indian language text classification tasks are far and few as most of the research employ traditional machine learning and deep learning models (Joshi et al., 2019; Tummalapalli  ... 
arXiv:2102.10275v1 fatcat:rt4nvzik3rdxhdzu34r55pjpoi

L3CubeMahaSent: A Marathi Tweet-based Sentiment Analysis Dataset [article]

Atharva Kulkarni, Meet Mandhane, Manali Likhitkar, Gayatri Kshirsagar, Raviraj Joshi
2021 arXiv   pre-print
Finally, we present the statistics of our dataset and baseline classification results using CNN, LSTM, ULMFiT, and BERT-based deep learning models.  ...  However, the Marathi language which is the third most popular language in India still lags behind due to the absence of proper datasets.  ...  We would like to express our gratitude towards our mentors at L3Cube for their continuous support and encouragement.  ... 
arXiv:2103.11408v2 fatcat:bycabla2mbfolonfcfmit5ezsu

Predicting Marathi News Class Using Semantic Entity-Driven Clustering Approach

Jatinderkumar R. Saini, Prafulla Bharat Bafna
2021 Journal of Cases on Information Technology  
Marathi corpus consisting of news is processed to form Group Entity document matrix Marathi (GEDMM), Vector space model for Marathi (VSMM) and Hysynset Vector space model for Marathi (HSVSMM).  ...  A random forest classifier is applied and the results are evaluated using misclassification error and accuracy.  ...  Outcomes provide analysis based on deep learning models to enhance text classification accuracy for Arabic documents (Wahdan et al., 2020) Named entitiy recognition is popularly used in English language  ... 
doi:10.4018/jcit.20211001.oa12 fatcat:qhjgvrohgngoxowsftwj5ab4ta


Gita Sinha, Dr. Shailja Sharma, Rakesh Kumar Roshan
2020 International Journal of Engineering Applied Sciences and Technology  
In this paper, we present various classification methods for printedoptical character identification (POCR) similarly handwritten optical character identification (HOCR).  ...  This article illustrate analysis of previous paper, and also distinguish the most important once out of the diversity of superior existing classification and feature extraction techniques and we will standardize  ...  Neural Network ,Deep Belief Network, Deep Neural Network, Deep Extreme Learning Machine and Localized Deep Extreme Learning Machine.  ... 
doi:10.33564/ijeast.2020.v05i03.023 fatcat:fy46yg4eujgq3ll3622vmnxdje

Marathi Poem Classification using Machine Learning

2019 International journal of recent technology and engineering  
Machine learning algorithm SVM classifier is used for differencing the class of the poem. This system also enables the user to search the poem based on the poet name and poet type.  ...  For 341 poems of five categories 'Friend', 'Prem', 'Bhakti', 'Prerna' and 'Desh' accuracy achieved is 93.54%.  ...  "Marathi Poem Classification Using Machine Learning". II.  ... 
doi:10.35940/ijrte.b1761.078219 fatcat:qtfefp255ncwbnqacojabjhdke

Performance Comparison for Spam Detection in Social Media Using Deep Learning Algorithms

Rushali Deshmukh, Et. al.
2021 Turkish Journal of Computer and Mathematics Education  
Projected Experimental results show the efficiency of the projected approach from the point of view of accuracy, F1-score and response time.  ...  The model is braced with linguistics data in the illustration of the words with the assistance of knowledge-bases such as Word2vec and fast ext.  ...  Convolutional Neural network is the deep learning algorithm that addresses the accurate classification of the text messages as spam or ham. Literature Survey Gauri Jain et al.  ... 
doi:10.17762/turcomat.v12i1s.1609 fatcat:lacff6y3s5d5rh6umxaqkn4yda

Machine Learning Techniques for Sentiment Analysis of Code-Mixed and Switched Indian Social Media Text Corpus - A Comprehensive Review

Gazi Imtiyaz Ahmad, Jimmy Singla, Anis Ali, Aijaz Ahmad Reshi, Anas A. Salameh
2022 International Journal of Advanced Computer Science and Applications  
A comprehensive review of sentiment analysis for code-mixed and switched text corpus of Indian social media using machine learning (ML) approaches, based on recent research studies has been presented in  ...  Systems have been developed using ML techniques to predict the polarity of code-mixed text corpus and to fine tune the existing models to improve their performance.  ...  The authors of [49] used Facebook comments of Hindi-English code-mixed dataset provided by Trolling, Aggression & Cyber bullying-I (TRAC-I) and apply machine and deep learning models for the classification  ... 
doi:10.14569/ijacsa.2022.0130254 fatcat:43ub7ku5xjeqvcjkpxfutpqgqi

WordAlchemy: A transformer-based Reverse Dictionary [article]

Dr. Sunil B. Mane, Harshal Patil, Kanhaiya Madaswar, Pranav Sadavarte
2022 arXiv   pre-print
In this paper, we propose a transformer-based deep learning approach to tackle the limitations faced by the existing systems using the mT5 model.  ...  Reverse Dictionaries are useful for new language learners, anomia patients, and for solving common tip-of-the-tongue problems (lethologica).  ...  Evaluation Metrics Referring to previous work (Hill et al., 2016 [7] ), we have used 4 evaluation metrics for our model: the accuracy of the target word being present in the top 1/10/100 i.e Acc@1/10/  ... 
arXiv:2204.10181v1 fatcat:givgtpukxfflllokycvxylibse

A Document Classification using NLP and Recurrent Neural Network

2019 International Journal of Engineering and Advanced Technology  
Classification of various document models based on short text, metadata, heading levels these are the existing techniques which are introduced in literature survey.  ...  We proposed a new document classification method based on deep learning using NLP and machine learning approach.  ...  In [18] they used deep learning to Arabic keyphrases extraction introduced Bi-LSTM neural network model, used to extract keyphrases from Arabic text.  ... 
doi:10.35940/ijeat.f8087.088619 fatcat:jcelq73ovrb2bmhjjmpznyfhhi

Teacher Perception of Automatically Extracted Grammar Concepts for L2 Language Learning [article]

Aditi Chaudhary, Arun Sampath, Ashwin Sheshadri, Antonios Anastasopoulos, Graham Neubig
2022 arXiv   pre-print
Specifically, we extract descriptions from a natural text corpus that answer questions about morphosyntax (learning of word order, agreement, case marking, or word formation) and semantics (learning of  ...  Overall, teachers find the materials to be interesting as a reference material for their own lesson preparation or even for learner evaluation.  ...  We also thank Pruthwik Mishra and Dipti Misra from IIIT-Hyderabad for sharing the Marathi and Kannada treebanks for training the parser.  ... 
arXiv:2206.05154v1 fatcat:afq2f75jivhmlcmtnk24wbe2qy

Indian Language Identification using Deep Learning

Shubham Godbole, Vaishnavi Jadhav, Gajanan Birajdar, M.D. Patil, V.A. Vyawahare
2020 ITM Web of Conferences  
Classification accuracy of 98.86% was obtained using the proposed methodology.  ...  Endeavours to create language recognizable proof frameworks for Indian dialects have been very restricted because of the issue of speaker accessibility and language readability.  ...  The model used for CNN classification (Table 2) is as given below: The performance of the proposed model can be evaluated from the confusion matrix (Table 1) .  ... 
doi:10.1051/itmconf/20203201010 fatcat:5udq3vdwnzab5ebyajucguse3m

Multitask Finetuning for Improving Neural Machine Translation in Indian Languages [article]

Shaily Desai, Atharva Kshirsagar, Manisha Marathe
2021 arXiv   pre-print
Pretraining these models on language modeling tasks and finetuning them on downstream tasks such as Text Classification, Question Answering and Neural Machine Translation has consistently shown exemplary  ...  We conduct an empirical study on three language pairs, Marathi-Hindi, Marathi-English and Hindi-English, where we compare the multitask finetuning approach to the standard finetuning approach, for which  ...  evaluations for all the language pairs considered.  ... 
arXiv:2112.01742v1 fatcat:7ig5mjmc35h6vk52gbzmwm435y

Code Switched and Code Mixed Speech Recognition for Indic languages [article]

Harveen Singh Chadha, Priyanshi Shah, Ankur Dhuriya, Neeraj Chhimwal, Anirudh Gupta, Vivek Raghavan
2022 arXiv   pre-print
The decoding information from a multilingual model is used for language identification and then combined with monolingual models to get an improvement of 50% WER across languages.  ...  We compare the performance of end to end multilingual speech recognition system to the performance of monolingual models conditioned on language identification (LID).  ...  model training and model testing.  ... 
arXiv:2203.16578v2 fatcat:4clay3fjxzhyvil6koxdl6smre

HCR-Net: A deep learning based script independent handwritten character recognition network [article]

Vinod Kumar Chauhan, Sukhdeep Singh, Anuj Sharma
2022 arXiv   pre-print
In this paper, we have proposed a novel deep learning architecture which exploits transfer learning and image-augmentation for end-to-end learning for script independent handwritten character recognition  ...  On the other hand, deep learning has witnessed huge success in different areas of pattern recognition, including HCR, and provides end-to-end learning but it has been studied for specific scripts only.  ...  test-set because it was used for final evaluation and Dig-mnist was used for evaluation during the training.Table12presents comparative study of HCR-Net against the state-of-art results.  ... 
arXiv:2108.06663v2 fatcat:mxq6psh6gjf6xp74rr5ozegyki
« Previous Showing results 1 — 15 out of 295 results