Posted: September 7th, 2024

COVID-19 Question Classification in Arabic Language

I need a paragraph about Arabic language and it`s challenges , paragraph about NLP , paragraph about Question answering system ,paragraph about CORONA virus. a paragraph about Question answering with machine learning
Literature Review about Arabic in NLP and in Question answering system like the following example i need a detailed paragraph about their work for each what do in details what tools used their results like the following example :
(Some researchers have experimented with question
classification techniques in the Arabic language. Al-Chalabi
et al. [6] presented a rule-based Arabic question classifying
technique. Their technique relies on the Arabic (IW) where
each IW within a question represents a class, while questions
that do not use IW were neglected. The authors considered the
IWs; (كم – how much, how many, how far, and how long), (من
– who), (ما – what), (اين – where), (متى – when), (اي – which),
(كيف – how). They have proposed patterns for each class as
illustrated in table 1. Most of the patterns start with IW and
maybe followed by a noun phrase (NP) or a verb phrase (VP),
and any word format (WF) that will not affect the
classification process. The only IW that may start with a
978-1-7281-2882-5/19/$31.00 ©2019 IEEE
proposition (PP) is (اي – which). The IW (ما – what) is either
followed by (HOA – ھو) ,(HEA – ھي ,(or a NP. The experiment
was conducted on 200 questions, applied on context freegrammar and regular expressions written using NooJ tool.
Results showed a recall and precision of 93% and 100%
respectively.
TABLE I. QUESTION PATTERNS PROPOSED IN [6]
IW ANSWER TYPE CLASS PATTERN
كم) HOW MUCH,
HOW MANY,
HOW FAR, AND
HOW LONG)
Number IW NP VP WF
IW VP WF
من) WHO) Person/ Organization IW NP WF
ما) WHAT) Device/ Geographical
location/ Sports/
Organization/ Art/ Person
IW HOA NP WF
IW HEA NP WF
IW NP WF
اين) WHERE) Geographical location IW VP WF
متى) WHEN) Date IW VP WF
اي) WHICH) Number/ Geographical
location/ History/ Sports
PP IW NP WF
IW NP WF
كيف) HOW) Science IW VP WF
Al-Shawakfa [7] proposed a rule-based technique to
classify questions according to IWs. The examined IWs; (من –
who, whose), (متى – when), (اين – where), (ماذا ,ما – what,
,much how – كم) ,(is what – ما ھو ,ماھي) ,(what – ما ,مما) ,(which
how many), (لماذا – why), (اي – which), (كيف – how). The
classes assigned are; person, organization, temporal
expressions, location, product, event, object, device, sports,
art, thing, numeric expressions, reason, history, and science.
The question is tokenized, then classified according to a set of
patterns defined by the authors. )

please will plagiarized, strong language , the references from journals with high impact factor like IEEE
COVID-19 Question Classification in Arabic Language
Over the years, the Arabic language has been criticized for being associated with terrorism, especially in the west. Today, the Arabic language is considered the most important language, especially in COVID-19 research. Arabic language is an official language to twenty-two countries, where according to studies, there are about three hundred thousand million Arabic speakers around the world, where most of the speakers live in the middle east and the north of Africa (Lahbari, El Alaoui, and Zidani, 2018). On the other hand, the Arabic language is part of the united nations’ official language. Apart from being the most used language in most countries, the Arabic language is not commonly used in the United Kingdom; only one percent can speak Arabic. The Arabic language is of different forms, where different countries use different forms of the Arabic language. The language is almost similar to the Hebrew language, where most speakers use Arabic manuscripts, which consist of various calligraphies.
Since Arabic is a semantic language, the words used are constructed from basic roots through a pattern of three letters. The Arabic language is a direct translation of various messages which is complex to construct. The best part about the Arabic language is that the language produces the text’s real meaning, emotions, and depth. The Arabic language has many forms of words for came, and the word love, for instance, the word Alaaqa, Hawa, Shanghai, and hub as the most used word for love (Marie-Sainte,et,al.,2018). Compared to other languages, such as English, Arabic; language is spelled from right to left, where the language does not consider words such as “h.”
The Arabic language faces various challenges despite the language being applied in science and art. The language cannot be used on the internet, especially in searching for various words. People using the language use English in searching for different words on the internet, such as in Google and other search webs (Ahmed,et, al., 2020). Another challenge is that Arabic cannot be used in most homes, schools, and organizations. The language can only be used in different events and for culture in Arabic countries. The Arabic language does not enhance literacy in Arabic countries. Most Arabic countries are illiterate, where people find it challenging to keep up with other developed countries NLP is considered a powerful language technique that assists individuals’ control, minds, and sound decisions (Lahbari, El Alaoui, and Zidani, 2018). Natural language processing (NLP) involves natural languages, such as software, speech, and texts. NLP language is fifty years old, where the language has been considered the most challenging to work with. Some of the forms of natural language used in everyday life include signs, emails, webpages, SMS, and menus (Nakov, Màrquez, Moschitti, and Mubarak, 2019). On the other hand, Arabic academic institutes have failed to make the language the main language used in communication. Most countries use the language to control, influence, and gain power over other countries.
NLP is considered a powerful language technique that assists individuals’ control, minds, and sound decisions. Natural language processing (NLP) involves natural languages, such as software, speech, and texts. NLP language is fifty years old, where the language has been considered the most challenging to work with. Some of the forms of natural language used in everyday life include signs, emails, webpages, SMS, and menus (Nakov, Màrquez, Moschitti, Mubarak, 2019).
Neurolinguistic programming (NPL) is a language that was developed in 1970 from Santa Cruze. NLP involves an analysis of various strategies that assist in attaining a goal through language, thoughts, and various behavioral experiences. As a psychological approach, neurolinguistic programming uses various logical levels of change, such as identity, beliefs, values, purpose, spirituality, and environment (Nakov, Màrquez, Moschitti, Mubarak, 2019). Neuro-linguistic programming helps change the way people think, especially on past events, and the approach to life. Neurolinguistic programming assists on how to control what goes around a person’s mind.
A question answering system is used to provide various responses to people’s questions through natural language. Various approaches to question answering questions involved the nature of the generated answer, the databases’ features, and the users’ queries. The current question answering system strives to meet various future needs, such as addressing critical issues, such as question classes, question processing, question answering, multi-lingual question answering, and interactive question and answer.
After the announcement of the COVID-19 outbreak in march 2020, Twitter and other social media platforms include millions of Arabic tweets concerning COVI-19. The tweeter involved a discussion on the virus; such s the outcome and the impacts of the disease in most countries (Verspoor,et, al., 2020). According to research studies, machine learning techniques have been used in identifying the nature of the tweets, which provide eighty-nine percent accuracy (Wahdan, Hantoobi, Salloum, and Shaalan, 2020). Through naïve Bayes, logistic regression, word frequency approach, word embedding, and support vector classification, the data were categorized into academic, media, health, and public data.
Researchers have conducted a study through natural language processing to understand the role of artificial intelligence and COVID-19. Since the Arabic language is used by around three hundred to four hundred and sixty-seven million people, most Arabs do not understand English twitters concerning COVID-19. Through the Arabic language, researchers have been able to get COVID-19 rumors and fake news. The tweeter feeds have been analyzed through the Arabic NLP techniques, where most scholars have updated the Arabic infectious disease ontology. According to Arabic translations of various COVID-19 rumors, children are not infected by corona, pets and mosquitos are transporters of coronavirus, extreme weather conditions can kill the virus, and various herbs protect people against coronavirus (Verspoor,et, al., 2020).

Question and answering in NLP is a broad concept, that involves the applications of machine learning (Ahmed, Ahmed, and Babu, 2017). Questions and answering techniques are used in machine learning, such as use of NLP techniques, such as conference resolution and parsing, and in creating chatbots and dialog systems (Ahmed, Hasan, Ali, and Mohammed, 2019). QA NLP has been used in neural networks (RNNs), memory networks, and attention mechanisms, enhancing the performance of machine learning-based question and answer. Several various neural network models have been applied to NLP, such as the RUS and the LSTMs used in summarization and classification of texts. Another deep learning technique used in solving QA tasks involves using the Word2 Vec model, which involves mapping words into usable data structures known as the word vectors (Ahmed, Ahmed, and Babu, 2017). The bab1 has been the most used dataset in training deep neural network models, which is divided into various files according to the type of question, for instance, the two or three supporting facts, the yes/no question, time reasoning, pathfinding, agents’ motivation, positional and size reasoning, and single supporting facts.
In understanding machines, the dataset is evaluated into various tasks, where some tasks are in English, while others in Hindi. The babi dataset, however, should be converted into other useful data structures for computation. For instance, the babi dataset can be split into questions, answers, or stories used by various models. Also, the dataset can be tokenized and combining with other stories. Lastly, the dataset can be indexing the questions and stories based on time of occurrence (Ahmed, Hasan, Ali, and Mohammed, 2019). However, the dataset is used to implement LTSM, memory networks, and dynamic memory networks.
Andreas, Rohrbach, Darrell, and Klein, 2016). Assignment help – Discuss the question answering that uses natural language in the neural network through various models used to answer various questions to neural networks. After translating the questions to neural networks, the networks are applied in word representation, such as through knowledge bases to produce answers. On the other hand, the authors provide a deep understanding of deep neural models’ impacts on captioning and image recognition (Ahmed, Ahmed, and Babu, 2017). The paper uses a model that explains both the continuous representation and linguistic compositionality of neural networks. However, the article further explains how deep learning is a functional program, such as the use of decomposing visual question answering. The authors use the dynamic neural module network in answering questions, which produce state-of-art results. For instance, the module assists in extending compositional question-answering machinery into complex and in the production of continuous world representations, such as images (Ahmed, Hasan, Ali, and Mohammed, 2019). On the other hand, the semantic structure prediction assists in a deep network.
Azmi, A. M., & Alshenaifi, N. A. (2017). Assignment help – Discuss the question answering systems that respond to questions that mostly consist of who and why. The article explains the use of LEMAZA, which is an Arabic name for why used to answer why questions. Additionally, LEMAZA why answering technique uses various forms that use rhetoric questions theory since most rhetoric questions begin with why.
Lahbari, Alaoui, and Zidani (2018) begin by discussing the importance of the question and answering technique as the fastest and easiest way of retrieving requested information, especially those written in NLP. The question answering system’s main goal is to answer questions effectively through the use of natural language processing and information retrieval. The paper discussed the English-Russian translator in 1954; the first question answering system used to handle two hundred and fifty words. Apart from NLP’s use as part of the research model, the method is also used in speech tagging, information retrieval, and named entity recognition. The authors discuss the Arabic language’s nature, as the most used language with ten thousand roots. The Arabic question classification includes question preprocessing, information retrieval, and answer processing (Wahdan, Hantoobi, Salloum, and Shaalan, 2020). The paper discusses several Arabic QAS, for instance, the QARAB, which was developed in 2002, DefArabia, introduces in 2010, the QArab Pro developed in 2011 as the most-used Arabic QAS.
On the other hand, the paper explains the features and processes involved in each question classification (Hamza, En-Nahnahi, Zidani, and Ouatik, 2019). For instance, in the preprocessing phase, tokenization of questions occurs, removing stop words, and removing punctuations. In classifying the questions, machine learning techniques, such as the support vector machine, and the decision tree, are used (Hamza, En-Nahnahi, Zidani, and Ouatik, 2019). On the other hand, before classification, various taxonomy types, such as the Arabic taxonomy, are the most used in Arabic text classification. The classification process uses features, such as the N-gram, the Bag-of -words, and the part-of-speech.

Ahmed, Ahmed, and Babu (2017) discuss the web-based Arabic question answering system using machine learning. The authors begin by discussing the importance of question answering systems as key tools in searching for answers to natural languages compared to most search engines. The author also discusses the key issues in question and answering systems, such as question processing, question classification, answers classification, and knowledge sources. The main approaches used in question answering systems, according to Ahmed, Ahmed, and Babu, include the statistical-based method, linguistic-based approach, and pattern matching approach. The commonly used linguistic resources and tools include the part-of-speech taggers, the parsers, the morphological analyzers, and the wordnets.
The approaches are not so accurate, especially in performance, where machine learning has been a better alternative, especially in considering the language used and the language pattern. The machine learning approach uses support vector machine SVM for question classification and stemming algorithm in removing suffixes; for instance, the author uses the Shereen Khoja stemming. In question extraction, the author uses the syntactical and lexical features through various feature spaces, such as the trigram, Bigram, and the Wh-Word.
Al-Shawakfa (2016) discusses the complexity of natural language processing (NLP), especially the lack of focus and interest among the Arabic researchers. On the other hand, the author discusses the challenges encountered in question answering systems, such as lack of proper answers through tagging rules, question analysis rules, and the 60 plus tagging rules. Although despite the challenges, the question answering systems provide accuracy is seventy-eight percent, and the recall ability is ninety-seven percent. On the other hand, the paper discusses a question answering system developed to improve the system’s results. The method is known as the IDRAAQ, which uses distance density N-gram, and query expansion to enhance the results’ performance. Also, AL QASIM was a model used to answer multiple questions and provides accurate results compared to the IDRAAQ. The paper also discusses other QA systems such as the EWAQ, whose accuracy is beyond Google, yahoo, and the factoid QA. The article mentions name entity recognition (NER) systems, rule-based approach, and machine learning (ML) approach (Wahdan, Hantoobi, Salloum, and Shaalan, 2020).
Samy, Hassanein, and Shaalan are authors that discuss Arabic question answering, which includes the challenges, systems, and techniques. According to the author, the question answering technique is computer systems’ ability to answer natural language questions, through question processing, answers processing, and documents processing, as the main module. According to the author, question processing is a min module that involves the generation of syntactic, and question analysis, and other sub-classifications, such as keyword extraction, and answer type generation, and keyword extraction.
According to the author, question classification involves categorizing the question into classes, such as factoid, Boolean, and other classes, such as a list. The author further discusses the properties of a question, the difference between question answering, database querying, and various types of QAS. The types of QAS includes the QA systems based on NLP and those that reason with NLP. Also, the author winds up by discussing the key components of QAS, which includes question processing, Arabic QAS on unstructured data, and architecture on linked data. Lastly, Samy, Hassanein, and Shaalan provide the challenges of Arabic QAS, such as lack of free word order and linguistic resources.

References
Ahmed, I. H., Nidhoimi, S. O. A., Badrasawi, K. J., & Mamat, A. (2020). CHALLENGES FACING TEACHERS OF ARABIC LANGUAGE (A2 PROGRAM) IN COMOROS SECONDARY SCHOOLS IN DEVELOPING 21ST CENTURY LEARNING SKILLS FOR ARABIC LANGUAGE STUDENTS. International E-Journal of Advances in Social Sciences, 5(15), 1659-1670.
Ahmed, M. A., Hasan, R. A., Ali, A. H., & Mohammed, M. A. (2019). The classification of the modern arabic poetry using machine learning. Telkomnika, 17(5), 2667-2674.
Ahmed, W., Ahmed, A., & Babu, A. P. (2017). Web-based Arabic question answering system using machine learning approach. International Journal of Advanced Research in Computer Science, 8(1).
Ahmed, W., Dasan, A., & Babu, A. P. (2017). Developing an Intelligent Question Answering System. International Journal of Education and Management Engineering, 7(6), 50.
Alotaibi, S. S. (2015). Sentiment analysis in the Arabic language using machine learning. 2000-2019-CSU Theses and Dissertations.
Alsayat, A., & Elmitwally, N. (2020). A comprehensive study for Arabic Sentiment Analysis (Challenges and Applications). Egyptian Informatics Journal, 21(1), 7-12.
Al-Shawakfa, E. (2016). ARule-BASED APPROACH TO UNDERSTAND QUESTIONS IN ARABIC QUESTION ANSWERING. Jordanian Journal of Computers and Information Technology (JJCIT), 2(3), 210-231.
AlShenaifi, N., & Azmi, A. (2020, December). Faheem at NADI shared task: Identifying the dialect of Arabic tweet. In Proceedings of the Fifth Arabic Natural Language Processing Workshop (pp. 282-287).
Al-Smadi, M., Al-Zboon, S., Jararweh, Y., & Juola, P. (2020). Transfer Learning for Arabic Named Entity Recognition With Deep Neural Networks. IEEE Access, 8, 37736-37745.
Alsudias, L., & Rayson, P. (2020, July). COVID-19 and Arabic Twitter: How can Arab World Governments and Public Health Organizations Learn from Social Media?. In Proceedings of the 1st Workshop on NLP for COVID-19 at ACL 2020.
Alyafeai, Z., & Al-Shaibani, M. (2020, November). ARBML: Democritizing Arabic Natural Language Processing Tools. In Proceedings of Second Workshop for NLP Open Source Software (NLP-OSS) (pp. 8-13).
Ameur, M. S. H., Meziane, F., & Guessoum, A. (2020). Arabic Machine Translation: A survey of the latest trends and challenges. Computer Science Review, 38, 100305.
Ameur, M. S. H., Meziane, F., & Guessoum, A. (2020). Arabic Machine Translation: A survey of the latest trends and challenges. Computer Science Review, 38, 100305.
Andreas, J., Rohrbach, M., Darrell, T., & Klein, D. (2016). Learning to compose neural networks for question answering. arXiv preprint arXiv:1601.01705.
Azmi, A. M., & Alshenaifi, N. A. (2017). Lemaza: An Arabic why-question answering system. Natural Language Engineering, 23(6), 877-903.
Dhaou, G., & Lejeune, G. (2020, December). Comparison between Voting Classifier and Deep Learning methods for Arabic Dialect Identification. In Proceedings of the Fifth Arabic Natural Language Processing Workshop (pp. 243-249).
Duwairi, R. M. (2006). Machine learning for Arabic text categorization. Journal of the American Society for Information Science and Technology, 57(8), 1005-1010.
Elarnaoty, M., AbdelRahman, S., & Fahmy, A. (2012). A machine learning approach for opinion holder extraction in Arabic language. arXiv preprint arXiv:1206.1011.
Haddad, B., Orabe, Z., Al-Abood, A., & Ghneim, N. (2020, May). Arabic Offensive Language Detection with Attention-based Deep Neural Networks. In Proceedings of the 4th Workshop on Open-Source Arabic Corpora and Processing Tools, with a Shared Task on Offensive Language Detection (pp. 76-81).
Hammad, M., & Al-awadi, M. (2016). Sentiment analysis for arabic reviews in social networks using machine learning. In Information technology: new generations (pp. 131-139). Springer, Cham.
Hamza, A., En-Nahnahi, N., & Ouatik, S. E. A. (2020, April). Exploring Contextual word representation for Arabic question classification. In 2020 1st International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET) (pp. 1-5). IEEE.
Hamza, A., En-Nahnahi, N., Ezziyyani, M., & Ouatik, S. E. A. (2019, July). Exploring Convolutional Neural Networks and Recurrent Neural Networks for Arabic Question Classification. In International Conference on Advanced Intelligent Systems for Sustainable Development (pp. 233-242). Springer, Cham.
Hamza, A., En-Nahnahi, N., Ezziyyani, M., & Ouatik, S. E. A. (2019, July). Exploring Convolutional Neural Networks and Recurrent Neural Networks for Arabic Question Classification. In International Conference on Advanced Intelligent Systems for Sustainable Development (pp. 233-242). Springer, Cham.
Hamza, A., En-Nahnahi, N., Zidani, K. A., & Ouatik, S. E. A. (2019). An arabic question classification method based on new taxonomy and continuous distributed representation of words. Journal of King Saud University-Computer and Information Sciences.
Jelodar, H., Wang, Y., Orji, R., & Huang, H. (2020). Deep sentiment classification and topic discovery on novel coronavirus or covid-19 online discussions: Nlp using lstm recurrent neural network approach. arXiv preprint arXiv:2004.11695.
Kanan, T., Sadaqa, O., Aldajeh, A., Alshwabka, H., AlZu’bi, S., Elbes, M., … & Alia, M. A. (2019, April). A review of natural language processing and machine learning tools used to analyze arabic social media. In 2019 IEEE Jordan International Joint Conference on Electrical Engineering and Information Technology (JEEIT) (pp. 622-628). IEEE.
Khalatia, M. M., & Al-Romanyb, T. A. H. (2020). Artificial Intelligence Development and Challenges (Arabic Language as a Model). Artificial Intelligence, 13(5).
Lahbari, I., El Alaoui, S. O., & Zidani, K. A. (2018). Toward a new arabic question answering system. Int. Arab J. Inf. Technol., 15(3A), 610-619.
Mahadi, T. S. T. Translation Challenges of Arabic Built-in-Language Repetition into English.
Marie-Sainte, S. L., Alalyani, N., Alotaibi, S., Ghouzali, S., & Abunadi, I. (2018). Arabic natural language processing and machine learning-based systems. IEEE Access, 7, 7011-7020.
MOY’AWIAH, A., NAHAR, K. M., & HALAWANI, K. M. H. (2019). AQAS: ARABIC QUESTION ANSWERING SYSTEM BASED ON SVM, SVD, and LSI. Journal of Theoretical and Applied Information Technology, 97(2).
Nakov, P., Màrquez, L., Moschitti, A., & Mubarak, H. (2019). Arabic community question answering. Natural Language Engineering, 25(1), 5-41.
Samy, H., Hassanein, E. E., & Shaalan, K. Arabic Question Answering: A Study on Challenges, Systems, and Techniques. International Journal of Computer Applications, 975, 8887.
Shaalan, K., Siddiqui, S., Alkhatib, M., & Monem, A. A. (2019). Challenges in arabic natural language processing. Computational Linguistics.
Verspoor, K., Cohen, K. B., Dredze, M., Ferrara, E., May, J., Munro, R., … & Wallace, B. C. (2020, July). Proceedings of the 1st Workshop on NLP for COVID-19 at ACL 2020. In Proceedings of the 1st Workshop on NLP for COVID-19 at ACL 2020.
Wahdan, K. A., Hantoobi, S., Salloum, S. A., & Shaalan, K. (2020). A systematic review of text classification research based ondeep learning models in Arabic language. Int. J. Electr. Comput. Eng, 10(6), 6629-6643.
Wahdan, K. A., Hantoobi, S., Salloum, S. A., & Shaalan, K. (2020). A systematic review of text classification research based ondeep learning models in Arabic language. Int. J. Electr. Comput. Eng, 10(6), 6629-6643.

Order | Check Discount

Tags: PSY Papers, Psych Research Paper Sample, Psychology Assignment, Psychology Dissertation Writing, Psychology research paper