Nisheeth Joshi

Pragya Katyayan, Nisheeth Joshi

Human mind is the palace of curious questions that seek answers. Computational resolution of this challenge is possible through Natural Language Processing techniques. Statistical techniques like machine learning and deep learning require a lot of data to train and despite that they fail to tap into the nuances of language. Such systems usually perform best on close-domain datasets. We have proposed development of a rule-based open-domain question-answering system which is capable of answering questions of any domain from a corresponding context passage. We have used 1000 questions from SQuAD 2.0 dataset for testing the developed system and it gives satisfactory results. In this paper, we have described the structure of the developed system and have analyzed the performance.

Tyler A. Chang, Catherine Arnett, Abdelrahman Eldesokey et al. · uw

To date, there exist almost no culturally-specific evaluation benchmarks for large language models (LLMs) that cover a large number of languages and cultures. In this paper, we present Global PIQA, a participatory commonsense reasoning benchmark for over 100 languages, constructed by hand by 335 researchers from 65 countries around the world. The 116 language varieties in Global PIQA cover five continents, 14 language families, and 23 writing systems. In the non-parallel split of Global PIQA, over 50% of examples reference local foods, customs, traditions, or other culturally-specific elements. We find that state-of-the-art LLMs perform well on Global PIQA in aggregate, but they exhibit weaker performance in lower-resource languages (up to a 37% accuracy gap, despite random chance at 50%). Open models generally perform worse than proprietary models. Global PIQA highlights that in many languages and cultures, everyday knowledge remains an area for improvement, alongside more widely-discussed capabilities such as complex reasoning and expert knowledge. Beyond its uses for LLM evaluation, we hope that Global PIQA provides a glimpse into the wide diversity of cultures in which human language is embedded.

Nisheeth Joshi, Pragya Katyayan, Palak Arora

Machine Translation (MT) Evaluation is an integral part of the MT development life cycle. Without analyzing the outputs of MT engines, it is impossible to evaluate the performance of an MT system. Through experiments, it has been identified that what works for English and other European languages does not work well with Indian languages. Thus, In this paper, we have introduced a reference-based MT evaluation metric for Gujarati which is based on supervised learning. We have trained two versions of the metric which uses 25 features for training. Among the two models, one model is trained using 6 hidden layers with 500 epochs while the other model is trained using 10 hidden layers with 500 epochs. To test the performance of the metric, we collected 1000 MT outputs of seven MT systems. These MT engine outputs were compared with 1 human reference translation. While comparing the developed metrics with other available metrics, it was found that the metrics produced better human correlations.

Pragya Katyayan, Nisheeth Joshi

Question Answering (QA) has proved to be an arduous challenge in the area of natural language processing (NLP) and artificial intelligence (AI). Many attempts have been made to develop complete solutions for QA as well as improving significant sub-modules of the QA systems to improve the overall performance through the course of time. Questions are the most important piece of QA, because knowing the question is equivalent to knowing what counts as an answer (Harrah in Philos Sci, 1961 [1]). In this work, we have attempted to understand questions in a better way by using Quantum Machine Learning (QML). The properties of Quantum Computing (QC) have enabled classically intractable data processing. So, in this paper, we have performed question classification on questions from two classes of SelQA (Selection-based Question Answering) dataset using quantum-based classifier algorithms-quantum support vector machine (QSVM) and variational quantum classifier (VQC) from Qiskit (Quantum Information Science toolKIT) for Python. We perform classification with both classifiers in almost similar environments and study the effects of circuit depths while comparing the results of both classifiers. We also use these classification results with our own rule-based QA system and observe significant performance improvement. Hence, this experiment has helped in improving the quality of QA in general.

Shivani Singh Rathore, Bharti Nathani, Nisheeth Joshi et al.

In this paper, we have shown a script conversion (transliteration) technique that converts Sindhi text in the Devanagari script to the Perso-Arabic script. We showed this by incorporating a hybrid approach where some part of the text is converted using a rule base and in case an ambiguity arises then a probabilistic model is used to resolve the same. Using this approach, the system achieved an overall accuracy of 99.64%.

Radhika Sharma, Pragya Katyayan, Nisheeth Joshi

In this paper, we have shown a method of improving the quality of neural machine translation by translating/transliterating name entities as a preprocessing step. Through experiments we have shown the performance gain of our system. For evaluation we considered three types of name entities viz person names, location names and organization names. The system was able to correctly translate mostly all the name entities. For person names the accuracy was 99.86%, for location names the accuracy was 99.63% and for organization names the accuracy was 99.05%. Overall, the accuracy of the system was 99.52%

Nisheeth Joshi, Pragya Katyayan

People who are visually impaired face a lot of difficulties while studying. One of the major causes to this is lack of available text in Bharti Braille script. In this paper, we have suggested a scheme to convert text in major Indian languages into Bharti Braille. The system uses a hybrid approach where at first the text in Indian language is given to a rule based system and in case if there is any ambiguity then it is resolved by applying a LSTM based model. The developed model has also been tested and found to have produced near accurate results.

Nisheeth Joshi, Pragya Katyayan

In this paper, we have shown the improvement of English to Bharti Braille machine translation system. We have shown how we can improve a baseline NMT model by adding some linguistic knowledge to it. This was done for five language pairs where English sentences were translated into five Indian languages and then subsequently to corresponding Bharti Braille. This has been demonstrated by adding a sub-module for translating multi-word expressions. The approach shows promising results as across language pairs, we could see improvement in the quality of NMT outputs. The least improvement was observed in English-Nepali language pair with 22.08% and the most improvement was observed in the English-Hindi language pair with 23.30%.

Nisheeth Joshi, Pragya Katyayan, Syed Afroz Ahmed

In this paper, we have studied the performance and role of local optimizers in quantum variational circuits. We studied the performance of the two most popular optimizers and compared their results with some popular classical machine learning algorithms. The classical algorithms we used in our study are support vector machine (SVM), gradient boosting (GB), and random forest (RF). These were compared with a variational quantum classifier (VQC) using two sets of local optimizers viz AQGD and COBYLA. For experimenting with VQC, IBM Quantum Experience and IBM Qiskit was used while for classical machine learning models, sci-kit learn was used. The results show that machine learning on noisy immediate scale quantum machines can produce comparable results as on classical machines. For our experiments, we have used a popular restaurant sentiment analysis dataset. The extracted features from this dataset and then after applying PCA reduced the feature set into 5 features. Quantum ML models were trained using 100 epochs and 150 epochs on using EfficientSU2 variational circuit. Overall, four Quantum ML models were trained and three Classical ML models were trained. The performance of the trained models was evaluated using standard evaluation measures viz, Accuracy, Precision, Recall, F-Score. In all the cases AQGD optimizer-based model with 100 Epochs performed better than all other models. It produced an accuracy of 77% and an F-Score of 0.785 which were highest across all the trained models.

Shruti Tyagi, Deepti Chopra, Iti Mathur et al.

Machine Translation is one of the research fields of Computational Linguistics. The objective of many MT Researchers is to develop an MT System that produce good quality and high accuracy output translations and which also covers maximum language pairs. As internet and Globalization is increasing day by day, we need a way that improves the quality of translation. For this reason, we have developed a Classifier based Text Simplification Model for English-Hindi Machine Translation Systems. We have used support vector machines and Naïve Bayes Classifier to develop this model. We have also evaluated the performance of these classifiers.

Pooja Gupta, Nisheeth Joshi, Iti Mathur

Machine Translation is the challenging problem for Indian languages. Every day we can see some machine translators being developed, but getting a high quality automatic translation is still a very distant dream . The correct translated sentence for Hindi language is rarely found. In this paper, we are emphasizing on English-Hindi language pair, so in order to preserve the correct MT output we present a ranking system, which employs some machine learning techniques and morphological features. In ranking no human intervention is required. We have also validated our results by comparing it with human ranking.

Iti Mathur, Nisheeth Joshi, Hemant Darbari et al.

With the web getting bigger and assimilating knowledge about different concepts and domains, it is becoming very difficult for simple database driven applications to capture the data for a domain. Thus developers have come out with ontology based systems which can store large amount of information and can apply reasoning and produce timely information. Thus facilitating effective knowledge management. Though this approach has made our lives easier, but at the same time has given rise to another problem. Two different ontologies assimilating same knowledge tend to use different terms for the same concepts. This creates confusion among knowledge engineers and workers, as they do not know which is a better term then the other. Thus we need to merge ontologies working on same domain so that the engineers can develop a better application over it. This paper shows the development of one such matcher which merges the concepts available in two ontologies at two levels; 1) at string level and 2) at semantic level; thus producing better merged ontologies. We have used a graph matching technique which works at the core of the system. We have also evaluated the system and have tested its performance with its predecessor which works only on string matching. Thus current approach produces better results.

Iti Mathur, Nisheeth Joshi, Hemant Darbari et al.

Since long, corporations are looking for knowledge sources which can provide structured description of data and can focus on meaning and shared understanding. Structures which can facilitate open world assumptions and can be flexible enough to incorporate and recognize more than one name for an entity. A source whose major purpose is to facilitate human communication and interoperability. Clearly, databases fail to provide these features and ontologies have emerged as an alternative choice, but corporations working on same domain tend to make different ontologies. The problem occurs when they want to share their data/knowledge. Thus we need tools to merge ontologies into one. This task is termed as ontology matching. This is an emerging area and still we have to go a long way in having an ideal matcher which can produce good results. In this paper we have shown a framework to matching ontologies using graphs.

Rashmi Gupta, Nisheeth Joshi, Iti Mathur

In this paper we present an approach for estimating the quality of machine translation system. There are various methods for estimating the quality of output sentences, but in this paper we focus on Naïve Bayes classifier to build model using features which are extracted from the input sentences. These features are used for finding the likelihood of each of the sentences of the training data which are then further used for determining the scores of the test data. On the basis of these scores we determine the class labels of the test data.

Pooja Gupta, Nisheeth Joshi, Iti Mathur

Since long, research on machine translation has been ongoing. Still, we do not get good translations from MT engines so developed. Manual ranking of these outputs tends to be very time consuming and expensive. Identifying which one is better or worse than the others is a very taxing task. In this paper, we show an approach which can provide automatic ranks to MT outputs (translations) taken from different MT Engines and which is based on N-gram approximations. We provide a solution where no human intervention is required for ranking systems. Further we also show the evaluations of our results which show equivalent results as that of human ranking.

Nisheeth Joshi, Iti Mathur, Hemant Darbari et al.

Machine translation evaluation is a very important activity in machine translation development. Automatic evaluation metrics proposed in literature are inadequate as they require one or more human reference translations to compare them with output produced by machine translation. This does not always give accurate results as a text can have several different translations. Human evaluation metrics, on the other hand, lacks inter-annotator agreement and repeatability. In this paper we have proposed a new human evaluation metric which addresses these issues. Moreover this metric also provides solid grounds for making sound assumptions on the quality of the text produced by a machine translation.

Vaishali Gupta, Nisheeth Joshi, Iti Mathur

Urdu is a combination of several languages like Arabic, Hindi, English, Turkish, Sanskrit etc. It has a complex and rich morphology. This is the reason why not much work has been done in Urdu language processing. Stemming is used to convert a word into its respective root form. In stemming, we separate the suffix and prefix from the word. It is useful in search engines, natural language processing and word processing, spell checkers, word parsing, word frequency and count studies. This paper presents a rule based stemmer for Urdu. The stemmer that we have discussed here is used in information retrieval. We have also evaluated our results by verifying it with a human expert.

Vaishali Gupta, Nisheeth Joshi, Iti Mathur

Machine translation is research based area where evaluation is very important phenomenon for checking the quality of MT output. The work is based on the evaluation of English to Urdu Machine translation. In this research work we have evaluated the translation quality of Urdu language which has been translated by using different Machine Translation systems like Google, Babylon and Ijunoon. The evaluation process is done by using two approaches - Human evaluation and Automatic evaluation. We have worked for both the approaches where in human evaluation emphasis is given to scales and parameters while in automatic evaluation emphasis is given to some automatic metric such as BLEU, GTM, METEOR and ATEC.

Jyoti Singh, Nisheeth Joshi, Iti Mathur

Part-of-speech (POS) tagging is a process of assigning the words in a text corresponding to a particular part of speech. A fundamental version of POS tagging is the identification of words as nouns, verbs, adjectives etc. For processing natural languages, Part of Speech tagging is a prominent tool. It is one of the simplest as well as most constant and statistical model for many NLP applications. POS Tagging is an initial stage of linguistics, text analysis like information retrieval, machine translator, text to speech synthesis, information extraction etc. In POS Tagging we assign a Part of Speech tag to each word in a sentence and literature. Various approaches have been proposed to implement POS taggers. In this paper we present a Marathi part of speech tagger. It is morphologically rich language. Marathi is spoken by the native people of Maharashtra. The general approach used for development of tagger is statistical using Unigram, Bigram, Trigram and HMM Methods. It presents a clear idea about all the algorithms with suitable examples. It also introduces a tag set for Marathi which can be used for tagging Marathi text. In this paper we have shown the development of the tagger as well as compared to check the accuracy of taggers output. The three Marathi POS taggers viz. Unigram, Bigram, Trigram and HMM gives the accuracy of 77.38%, 90.30%, 91.46% and 93.82% respectively.

Deepti Bhalla, Nisheeth Joshi, Iti Mathur

This paper presents a novel approach to machine translation by combining the state of art name entity translation scheme. Improper translation of name entities lapse the quality of machine translated output. In this work, name entities are transliterated by using statistical rule based approach. This paper describes the translation and transliteration of name entities from English to Punjabi. We have experimented on four types of name entities which are: Proper names, Location names, Organization names and miscellaneous. Various rules for the purpose of syllabification have been constructed. Transliteration of name entities is accomplished with the help of Probability calculation. N-Gram probabilities for the extracted syllables have been calculated using statistical machine translation toolkit MOSES.

Rashmi Gupta, Nisheeth Joshi, Iti Mathur

This paper considers the problem for estimating the quality of machine translation outputs which are independent of human intervention and are generally addressed using machine learning techniques.There are various measures through which a machine learns translations quality. Automatic Evaluation metrics produce good co-relation at corpus level but cannot produce the same results at the same segment or sentence level. In this paper 16 features are extracted from the input sentences and their translations and a quality score is obtained based on Bayesian inference produced from training data.

Nisheeth Joshi, Hemant Darbari, Iti Mathur

For the past 60 years, Research in machine translation is going on. For the development in this field, a lot of new techniques are being developed each day. As a result, we have witnessed development of many automatic machine translators. A manager of machine translation development project needs to know the performance increase/decrease, after changes have been done in his system. Due to this reason, a need for evaluation of machine translation systems was felt. In this article, we shall present the evaluation of some machine translators. This evaluation will be done by a human evaluator and by some automatic evaluation metrics, which will be done at sentence, document and system level. In the end we shall also discuss the comparison between the evaluations.

Jyoti Singh, Nisheeth Joshi, Iti Mathur

In this paper we present a Marathi part of speech tagger. It is a morphologically rich language. It is spoken by the native people of Maharashtra. The general approach used for development of tagger is statistical using trigram Method. The main concept of trigram is to explore the most likely POS for a token based on given information of previous two tags by calculating probabilities to determine which is the best sequence of a tag. In this paper we show the development of the tagger. Moreover we have also shown the evaluation done.

Deepti Bhalla, Nisheeth Joshi, Iti Mathur

Machine Transliteration has come out to be an emerging and a very important research area in the field of machine translation. Transliteration basically aims to preserve the phonological structure of words. Proper transliteration of name entities plays a very significant role in improving the quality of machine translation. In this paper we are doing machine transliteration for English-Punjabi language pair using rule based approach. We have constructed some rules for syllabification. Syllabification is the process to extract or separate the syllable from the words. In this we are calculating the probabilities for name entities (Proper names and location). For those words which do not come under the category of name entities, separate probabilities are being calculated by using relative frequency through a statistical machine translation toolkit known as MOSES. Using these probabilities we are transliterating our input text from English to Punjabi.

Juhi Ameta, Nisheeth Joshi, Iti Mathur

Machine Translation for Indian languages is an emerging research area. Transliteration is one such module that we design while designing a translation system. Transliteration means mapping of source language text into the target language. Simple mapping decreases the efficiency of overall translation system. We propose the use of stemming and part-of-speech tagging for transliteration. The effectiveness of translation can be improved if we use part-of-speech tagging and stemming assisted transliteration.We have shown that much of the content in Gujarati gets transliterated while being processed for translation to Hindi language.

Snigdha Paul, Nisheeth Joshi, Iti Mathur

We live in a translingual society, in order to communicate with people from different parts of the world we need to have an expertise in their respective languages. Learning all these languages is not at all possible; therefore we need a mechanism which can do this task for us. Machine translators have emerged as a tool which can perform this task. In order to develop a machine translator we need to develop several different rules. The very first module that comes in machine translation pipeline is morphological analysis. Stemming and lemmatization comes under morphological analysis. In this paper we have created a lemmatizer which generates rules for removing the affixes along with the addition of rules for creating a proper root word.

Nisheeth Joshi, Iti Mathur

Developing parallel corpora is an important and a difficult activity for Machine Translation. This requires manual annotation by Human Translators. Translating same text again is a useless activity. There are tools available to implement this for European Languages, but no such tool is available for Indian Languages. In this paper we present a tool for Indian Languages which not only provides automatic translations of the previously available translation but also provides multiple translations, in cases where a sentence has multiple translations, in ranked list of suggestive translations for a sentence. Moreover this tool also lets translators have global and local saving options of their work, so that they may share it with others, which further lightens the task.

Juhi Ameta, Nisheeth Joshi, Iti Mathur

Gujarati is a resource poor language with almost no language processing tools being available. In this paper we have shown an implementation of a rule based stemmer of Gujarati. We have shown the creation of rules for stemming and the richness in morphology that Gujarati possesses. We have also evaluated our results by verifying it with a human expert.

Archana Vashisth, Iti Mathur, Nisheeth Joshi

Today, we can find many search engines which provide us with information which is more operational in nature. None of the search engines provide domain specific information. This becomes very troublesome to a novice user who wishes to have information in a particular domain. In this paper, we have developed an ontology which can be used by a domain specific search engine. We have developed an ontology on human anatomy, which captures information regarding cardiovascular system, digestive system, skeleton and nervous system. This information can be used by people working in medical and health care domain.

Nisheeth Joshi, Iti Mathur

Written Communication on Computers requires knowledge of writing text for the desired language using Computer. Mostly people do not use any other language besides English. This creates a barrier. To resolve this issue we have developed a scheme to input text in Hindi using phonetic mapping scheme. Using this scheme we generate intermediate code strings and match them with pronunciations of input text. Our system show significant success over other input systems available.

Nisheeth Joshi, Iti Mathur

Natural Language Parsing has been the most prominent research area since the genesis of Natural Language Processing. Probabilistic Parsers are being developed to make the process of parser development much easier, accurate and fast. In Indian context, identification of which Computational Grammar Formalism is to be used is still a question which needs to be answered. In this paper we focus on this problem and try to analyze different formalisms for Indian languages.