Part-of-Speech (POS) Tagging Using Deep Learning-Based Approaches on the Designed Khasi POS Corpus

Part-of-speech (POS) tagging is one of the research challenging fields in natural language processing (NLP). It requires good knowledge of a particular language with large amounts of data or corpora for feature engineering, which can lead to achieving a good performance of the tagger. Our main contribution in this research work is the designed Khasi POS corpus. Till date, there has been no form of any kind of Khasi corpus developed or formally developed. In the present designed Khasi POS corpus, each word is tagged manually using the designed tagset. Methods of deep learning have been used to experiment with our designed Khasi POS corpus. The POS tagger based on BiLSTM, combinations of BiLSTM with CRF, and character-based embedding with BiLSTM are presented. The main challenges of understanding and handling Natural Language toward Computational linguistics to encounter are anticipated. In the presently designed corpus, we have tried to solve the problems of ambiguities of words concerning their context usage, and also the orthography problems that arise in the designed POS corpus. The designed Khasi corpus size is around 96,100 tokens and consists of 6,616 distinct words. Initially, while running the first few sets of data of around 41,000 tokens in our experiment the taggers are found to yield considerably accurate results. When the Khasi corpus size has been increased to 96,100 tokens, we see an increase in accuracy rate and the analyses are more pertinent. As results, accuracy of 96.81% is achieved for the BiLSTM method, 96.98% for BiLSTM with CRF technique, and 95.86% for character-based with LSTM. Concerning substantial research from the NLP perspectives for Khasi, we also present some of the recently existing POS taggers and other NLP works on the Khasi language for comparative purposes.

A Modified Markov Based Maximum-entropy Model for POS Tagging of Odia Text

POS (Parts of Speech) tagging, a vital step in diverse Natural Language Processing (NLP) tasks has not drawn much attention in case of Odia a computationally under-developed language. The proposed hybrid method suggests a robust POS tagger for Odia. Observing the rich morphology of the language and unavailability of sufficient annotated text corpus a combination of machine learning and linguistic rules is adopted in the building of the tagger. The tagger is trained on tagged text corpus from the domain of tourism and is capable of obtaining a perceptible improvement in the result. Also an appreciable performance is observed for news articles texts of varied domains. The performance of proposed algorithm experimenting on Odia language shows its manifestation in dominating over existing methods like rule based, hidden Markov model (HMM), maximum entropy (ME) and conditional random field (CRF).

A Hidden Markov Model-based Part of Speech Tagger for Shekki’noono Language

Natural language processing plays a great role in providing an interface for human-computer communication. It enables people to talk with the computer in their formal language rather than machine language. This study aims at presenting a Part of speech tagger that can assign word class to words in a given paragraph sentence. Some of the researchers developed parts of speech taggers for different languages such as English Amharic, Afan Oromo, Tigrigna, etc. On the other hand, many other languages do not have POS taggers like Shekki’noono language. POS tagger is incorporated in most natural language processing tools like machine translation, information extraction as a basic component. So, it is compulsory to develop a part of speech tagger for languages then it is possible to work with an advanced natural language application. Because those applications enhance machine to machine, machine to human, and human to human communications. Although, one language POS tagger cannot be directly applied for other languages POS tagger. With the purpose for developing the Shekki’noono POS tagger, we have used the stochastic Hidden Markov Model. For the study, we have used 1500 sentences collected from different sources such as newspapers (which includes social, economic, and political aspects), modules, textbooks, Radio Programs, and bulletins. The collected sentences are labeled by language experts with their appropriate parts of speech for each word. With the experiments carried out, the part of speech tagger is trained on the training sets using Hidden Markov model. As experiments showed, HMM based POS tagging has achieved 92.77 % accuracy for Shekki’noono. And the POS tagger model is compared with the previous experiments in related works using HMM. As a future work, the proposed approaches can be utilized to perform an evaluation on a larger corpus.

A Template-Based Approach for Tagging Non-Vocalized Arabic Nouns

There exist no corpora of Arabic nouns. Furthermore, in any Arabic text, nouns can be found in different forms. In fact, by tagging nouns in an Arabic text, the beginning of each sentence can determine whether it starts with a noun or a verb. Part of Speech Tagging (POS) is the task of labeling each word in a sentence with its appropriate category, which is called a Tag (Noun, Verb and Article). In this thesis, we attempt to tag non-vocalized Arabic text. The proposed POS Tagger for Arabic Text is based on searching for each word of the text in our lists of Verbs and Articles. Nouns are found by eliminating Verbs and Articles. Our hypothesis states that, if the word in the text is not found in our lists, then it is a Noun. These comparisons will be made for each of the words in the text until all of them have been tagged. To apply our method, we have prepared a list of articles and verbs in the Arabic language with a total of 112 million verbs and articles combined, which are used in our comparisons to prove our hypothesis. To evaluate our proposed method, we used pre-tagged words from "The Quranic Arabic Corpus", making a total of 78,245 words, with our method, the Template-based tagging approach compared with (AraMorph) a rule-based tagging approach and the Stanford Log-linear Part-Of-Speech Tagger. Finally, AraMorph produced 40% correctly-tagged words and Stanford Log-linear Part-Of-Speech Tagger produced 68% correctly-tagged words, while our method produced 68,501 correctly-tagged words (88%).

An HMM-Based PoS Tagger for Old Church Slavonic

We present a hybrid HMM-based PoS tagger for Old Church Slavonic. The training corpus is a portion of one text, Codex Marianus (40k) annotated with the Universal Dependencies UPOS tags in the UD-PROIEL treebank. We perform a number of experiments in within-domain and out-of-domain settings, in which the remaining part of Codex Marianus serves as a within-domain test set, and Kiev Folia is used as an out-of-domain test set. Analysing by-PoS-class precision and sensitivity in each run, we combine a simple context-free n-gram-based approach and Hidden Markov method (HMM), and added linguistic rules for specific cases such as punctuation and digits. While the model achieves a rather non-impressive accuracy of 81% in in-domain settings, we observe an accuracy of 51% in out-of-domain evaluation, which is comparable to the results of large neural architectures based on pre-trained contextual embeddings.

Universal Dependencies for Tweets in Brazilian Portuguese: Tokenization and Part of Speech Tagging

Automatically dealing with Natural Language User-Generated Content (UGC) is a challenging task of utmost importance, given the amount of information available over the web. We present in this paper an effort on building tokenization and Part of Speech (PoS) tagging systems for tweets in Brazilian Portuguese, following the guidelines of the Universal Dependencies (UD) project. We propose a rule-based tokenizer and the customization of current state-of-the-art UD-based tagging strategies for Portuguese, achieving a 98% f-score for tokenization, and a 95% f-score for PoS tagging. We also introduce DANTEStocks, the corpus of stock market tweets on which we base our work, presenting preliminary evidence of the multi-genre capacity of our PoS tagger.

POS Tagging and NER System for Kannada Using Conditional Random Fields

Parts-of-speech (POS) tagging is a method used to assign the POS tag for every word present in the text, and named entity recognition (NER) is a process to identify the proper nouns in the text and to classify the identified nouns into certain predefined categories. A POS tagger and a NER system for Kannada text have been proposed utilizing conditional random fields (CRFs). The dataset used for POS tagging consists of 147K tokens, where 103K tokens are used for training and the remaining tokens are used for testing. The proposed CRF model for POS tagging of Kannada text obtained 91.3% of precision, 91.6% of recall, and 91.4% of f-score values, respectively. To develop the NER system for Kannada, the data required is created manually using the modified tag-set containing 40 labels. The dataset used for NER system consists of 16.5K tokens, where 70% of the total words are used for training the model, and the remaining 30% of total words are used for model testing. The developed NER model obtained the 94% of precision, 93.9% of recall, and 93.9% of F1-measure values, respectively.

AUTOMATED FUNCTIONAL ANALYSIS OF PATENTS FOR PRODUCING DESIGN INSIGHT

Patent analysis is a popular topic of research. However, designers do not engage with patents in the early design stage, as patents are time-consuming to read and understand due to their intricate structure and the legal terminologies used. Manually produced graphical representations of patent working principles for improving designers’ awareness of prior art have been demonstrated in previous research. In this paper, an automated approach is presented, utilising Natural Language Processing (NLP) techniques to identify the invention working principle from the patent independent claims and produce a visualisation. The outcomes of this automated approach are compared with previous manually produced examples. The results indicate over 40% match between the automatic and manual approach, which is a good basis for further development. The comparison suggests that the automated approach works well for features and relationships that are expressed explicitly and consistently but begin to lose accuracy when applied to complex sentences. The comparison also suggests that the accuracy of the proposed automated approach can be improved by using a trained part-of-speech (POS) tagger, improved parsing grammar and an ontology.

An Efficient Soft Computing Approach for Text Identification using Artificial Intelligence Model

This paper presents an enhanced system in the field of text identification using Soft computing techniques. The model designed in this work analyzes the blogs or input text and classifies the personality into five major categories; Neuroticism, Extraversion, Openness, Conscientiousness and Agreeableness. The blog or text is first passed through POS tagger then a feature vector matrix is generated according to the attributes of the personality chart. Each column of FVM is calculated in its domain that improves the final result of personality identification. The result of the proposed model is improvement over similar work by other researchers [1, 2, 3].