Pub Date : 2022-10-12DOI: 10.1109/ESOLEC54569.2022.10009168
Lutfieh S. Al-homed, K. M. Jambi, Hassanin M. Al-Barhamtoshy
This paper presents a new dataset of Ancient Arabic-Islamic Manuscripts to detect unknown manuscripts and classify them from the known manuscripts. Unknown Manuscripts are identified as those that have been affected badly by human or natural forces, such as humidity, temperature, and air pollution, which degraded their quality and missed their identification information, such as the title, author, and date of the manuscripts. Thus, The Known Manuscripts are characterized by having a known title, author, etc. Recognizing the unknown manuscripts is essential to further the analysis process, facilitate information extraction from such degraded manuscripts, enable their indexing, and make them easily accessed and retrieved. The objectives of the constructed dataset are as follows: 1) Collect a set of known and unknown manuscripts of similar forms and highlight the characteristics of the unknown manuscripts. 2) Promote the automatic detection and recognition of unknown manuscripts. 3) Formulate the problem of recognizing unknown manuscripts as a supervised machine-learning problem, and boost this recognition with the advances in machine learning and deep learning techniques. A total of 108 manuscripts were collected, distributed equally by the known and unknown categories. The preliminary results for classifying and recognizing unknown manuscripts showed that using a decision tree classifier achieved an accuracy of 88% in classifying unknown manuscripts.
{"title":"A Novel Dataset for Known and Unknown Ancient Arabic Manuscripts","authors":"Lutfieh S. Al-homed, K. M. Jambi, Hassanin M. Al-Barhamtoshy","doi":"10.1109/ESOLEC54569.2022.10009168","DOIUrl":"https://doi.org/10.1109/ESOLEC54569.2022.10009168","url":null,"abstract":"This paper presents a new dataset of Ancient Arabic-Islamic Manuscripts to detect unknown manuscripts and classify them from the known manuscripts. Unknown Manuscripts are identified as those that have been affected badly by human or natural forces, such as humidity, temperature, and air pollution, which degraded their quality and missed their identification information, such as the title, author, and date of the manuscripts. Thus, The Known Manuscripts are characterized by having a known title, author, etc. Recognizing the unknown manuscripts is essential to further the analysis process, facilitate information extraction from such degraded manuscripts, enable their indexing, and make them easily accessed and retrieved. The objectives of the constructed dataset are as follows: 1) Collect a set of known and unknown manuscripts of similar forms and highlight the characteristics of the unknown manuscripts. 2) Promote the automatic detection and recognition of unknown manuscripts. 3) Formulate the problem of recognizing unknown manuscripts as a supervised machine-learning problem, and boost this recognition with the advances in machine learning and deep learning techniques. A total of 108 manuscripts were collected, distributed equally by the known and unknown categories. The preliminary results for classifying and recognizing unknown manuscripts showed that using a decision tree classifier achieved an accuracy of 88% in classifying unknown manuscripts.","PeriodicalId":179850,"journal":{"name":"2022 20th International Conference on Language Engineering (ESOLEC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113998902","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-12DOI: 10.1109/ESOLEC54569.2022.10009594
Ahmed H. Kassem, S. Alansary
This paper addresses the challenging task of identifying semantic features in the Quran from a corpus-based as well as computational perspective, namely color identification. The study attempts to identify, locate, and demonstrate the frequencies, occurrences, and concordances of the colors in the Quran using AntConc and The Simple Corpus Tool, the results are compared to earlier manual work and the information available at corpus.quran.com, a University of Leeds's Corpus work on the Holy Quran. The research undertakes the task of semantically annotating lexical items related to colors as well as examining them in concordance and corpus software tools. The results are compared with special attention to the colors' co-occurrences in an endeavor to better understand the connotations of colors in the Quran. The paper identifies a gap in the Leeds's Corpus work on the Quran and recommends filling the gap with the work entailed in the study.
本文从基于语料库和计算的角度解决了古兰经语义特征识别的挑战性任务，即颜色识别。该研究试图使用AntConc和The Simple Corpus Tool来识别、定位和展示古兰经中颜色的频率、出现频率和一致性，并将结果与早期的手工工作和corpus.quran.com上的信息进行比较，corpus.quran.com是利兹大学关于神圣古兰经的语料库工作。本研究的任务是对与颜色相关的词汇项目进行语义标注，并在语料库软件工具中对其进行检查。为了更好地理解《古兰经》中颜色的内涵，将这些结果与特别关注颜色的共现进行了比较。这篇论文指出了利兹的《古兰经》语料库工作中的一个空白，并建议用这项研究所涉及的工作来填补这一空白。
{"title":"Contrastive Analysis of Color Representations Using Semantic Corpus Annotation “POS Tagging”:The Holy Quran- A Case Study","authors":"Ahmed H. Kassem, S. Alansary","doi":"10.1109/ESOLEC54569.2022.10009594","DOIUrl":"https://doi.org/10.1109/ESOLEC54569.2022.10009594","url":null,"abstract":"This paper addresses the challenging task of identifying semantic features in the Quran from a corpus-based as well as computational perspective, namely color identification. The study attempts to identify, locate, and demonstrate the frequencies, occurrences, and concordances of the colors in the Quran using AntConc and The Simple Corpus Tool, the results are compared to earlier manual work and the information available at corpus.quran.com, a University of Leeds's Corpus work on the Holy Quran. The research undertakes the task of semantically annotating lexical items related to colors as well as examining them in concordance and corpus software tools. The results are compared with special attention to the colors' co-occurrences in an endeavor to better understand the connotations of colors in the Quran. The paper identifies a gap in the Leeds's Corpus work on the Quran and recommends filling the gap with the work entailed in the study.","PeriodicalId":179850,"journal":{"name":"2022 20th International Conference on Language Engineering (ESOLEC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128085952","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-12DOI: 10.1109/ESOLEC54569.2022.10009506
Mai Magdy M. Sleim
Cognition is mostly seen as the motivation towards certain emotive decisions and/or actions. Thus, it is only restricted to evaluation and appraisal in the studies of emotions. However, the long-term emotions, i.e. sentiments, are hardly represented in relation to cognitive aspects. The aim of this study is to provide an in-depth understanding of cognition and affection through an adaptation of Johnson's (1987) and Kimmel's (2005) image schema theory, and Plutchik's (1980, 1988) sentiment analysis in quotes of Goodreads. The data selected for this study consists of three themes of Goodreads: life, death, and inspiration. From each theme, six hundred quotes were selected and analyzed. The study focuses on the correlation between sentiments and image schemas detecting aspects of control and/or responsibility of the self and the other. Therefore, the third tool of analysis is used to identify sentiments, i.e. lexico-syntactic analysis which assists in the identification of the self/other control and responsibility through the agent/doer, experiencer knowledge. A mixed methodology, incorporating both qualitative and quantitative analyses, is adopted highlighting relationships between variables. The analysis is highlighted through the framework of Johnson's image schema theory and Plutchik's theories (Theory of Emotions and Theory of Cognition-Emotion Relations). The statistical analyses show that there is a significant relationship between cognition and emotions except with LOCOMOTION and SPACE. This might be due to the nature of those schemas where there are several subschemas in addition to their linear nature. This study adds texture to human knowledge as it detects different aspects of human experiences. Furthermore, the amount of manual data analyzed contributes to the fields of cognitive semantics and psychology, especially, image schema modeling.
{"title":"Sentiments and Cognition Interdependence: An Exploratory Study of Sentiment Analysis and Image Schema","authors":"Mai Magdy M. Sleim","doi":"10.1109/ESOLEC54569.2022.10009506","DOIUrl":"https://doi.org/10.1109/ESOLEC54569.2022.10009506","url":null,"abstract":"Cognition is mostly seen as the motivation towards certain emotive decisions and/or actions. Thus, it is only restricted to evaluation and appraisal in the studies of emotions. However, the long-term emotions, i.e. sentiments, are hardly represented in relation to cognitive aspects. The aim of this study is to provide an in-depth understanding of cognition and affection through an adaptation of Johnson's (1987) and Kimmel's (2005) image schema theory, and Plutchik's (1980, 1988) sentiment analysis in quotes of Goodreads. The data selected for this study consists of three themes of Goodreads: life, death, and inspiration. From each theme, six hundred quotes were selected and analyzed. The study focuses on the correlation between sentiments and image schemas detecting aspects of control and/or responsibility of the self and the other. Therefore, the third tool of analysis is used to identify sentiments, i.e. lexico-syntactic analysis which assists in the identification of the self/other control and responsibility through the agent/doer, experiencer knowledge. A mixed methodology, incorporating both qualitative and quantitative analyses, is adopted highlighting relationships between variables. The analysis is highlighted through the framework of Johnson's image schema theory and Plutchik's theories (Theory of Emotions and Theory of Cognition-Emotion Relations). The statistical analyses show that there is a significant relationship between cognition and emotions except with LOCOMOTION and SPACE. This might be due to the nature of those schemas where there are several subschemas in addition to their linear nature. This study adds texture to human knowledge as it detects different aspects of human experiences. Furthermore, the amount of manual data analyzed contributes to the fields of cognitive semantics and psychology, especially, image schema modeling.","PeriodicalId":179850,"journal":{"name":"2022 20th International Conference on Language Engineering (ESOLEC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125154223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-12DOI: 10.1109/ESOLEC54569.2022.10009437
H. Salama, S. Alansary, Amany Elshazly
The aim of this study is to provide a morphological analysis of the Egyptian children corpus, which is a morphologically tagged and disambiguated in CHILDES. This allows the KIDEVAL program to be readily used on the corpus to address questions regarding the acquisition of Egyptian Arabic. KIDEVAL is one of the useful tools in CLAN program which has been particularly useful toolsets in the study of language acquisition in many languages. However, applications of corpus-based analyses to Egyptian children's language have not yet been conducted. This study describes how to use the KIDEVAL program for analyzing Egyptian children's language and study the development of word frequency patterns of parts of speech and order of development of grammatical morphemes in Egyptian Arabic. The output of morphological analysis enables researchers to study and answer many questions regarding the development of a grammatical morpheme in Egyptian Arabic, as well as a lot of questions that can readily be probed with KIDEVAL. The Egyptian Arabic corpus is downloaded from the Arabic part of the CHILDES database. It comprises 10transcripts from Egyptian-speaking children aged 1;7 to3;8 years, with a total of 25,645 words. The KIDEVAL program analysis profile for Egyptian Arabic children's corpus in this study reveals extensive and valuable analysis, displaying the number of occurrences of each part of speech for each child depends on his age which includes 54 categories and subcategories. The usage of the KIDEVAL tool is efficient because it reduces the time needed to label the corpus manually.
{"title":"Morphological Analysis of Egyptian Children Corpus by KIDEVAL Program","authors":"H. Salama, S. Alansary, Amany Elshazly","doi":"10.1109/ESOLEC54569.2022.10009437","DOIUrl":"https://doi.org/10.1109/ESOLEC54569.2022.10009437","url":null,"abstract":"The aim of this study is to provide a morphological analysis of the Egyptian children corpus, which is a morphologically tagged and disambiguated in CHILDES. This allows the KIDEVAL program to be readily used on the corpus to address questions regarding the acquisition of Egyptian Arabic. KIDEVAL is one of the useful tools in CLAN program which has been particularly useful toolsets in the study of language acquisition in many languages. However, applications of corpus-based analyses to Egyptian children's language have not yet been conducted. This study describes how to use the KIDEVAL program for analyzing Egyptian children's language and study the development of word frequency patterns of parts of speech and order of development of grammatical morphemes in Egyptian Arabic. The output of morphological analysis enables researchers to study and answer many questions regarding the development of a grammatical morpheme in Egyptian Arabic, as well as a lot of questions that can readily be probed with KIDEVAL. The Egyptian Arabic corpus is downloaded from the Arabic part of the CHILDES database. It comprises 10transcripts from Egyptian-speaking children aged 1;7 to3;8 years, with a total of 25,645 words. The KIDEVAL program analysis profile for Egyptian Arabic children's corpus in this study reveals extensive and valuable analysis, displaying the number of occurrences of each part of speech for each child depends on his age which includes 54 categories and subcategories. The usage of the KIDEVAL tool is efficient because it reduces the time needed to label the corpus manually.","PeriodicalId":179850,"journal":{"name":"2022 20th International Conference on Language Engineering (ESOLEC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124747067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-12DOI: 10.1109/ESOLEC54569.2022.10009473
Alaa Elnily, Ahmed Abdelghany
The process of automatically giving the proper POS tag to each word in a text based on context is known as automatic POS tagging. The majority of NLP applications require this process as a crucial step. This study intends to propose a machine learning-based Arabic POS tagger. YAMCHA tool is the machine learning system employed in this study. YAMCHA utilizes Support Vector Machines as a machine learning algorithm. SVM classifies data with high accuracy because it makes use of part of data in training process. As a result, in order to train the system, a substantial amount of annotated data must be evaluated at the POS level. A corpus of 100,039 words is utilized in this study. It was divided into training and testing parts, totaling 64,608 and 35,431 words, respectively. A tag set of 48 morphological tags were used in training and testing. To reach the best result in the automatic POS tagging, the system was trained multiple times with changing the range of linguistic information used in training process, and then new texts were tested and evaluated. The least error rate achieved was 11.4%. This rate was reached when the preceding word of the target one was considered in the training process without considering its POS tag (F: −1‥0: 0‥).
{"title":"Automatic POS tagging of Arabic words using the YAMCHA machine learning tool","authors":"Alaa Elnily, Ahmed Abdelghany","doi":"10.1109/ESOLEC54569.2022.10009473","DOIUrl":"https://doi.org/10.1109/ESOLEC54569.2022.10009473","url":null,"abstract":"The process of automatically giving the proper POS tag to each word in a text based on context is known as automatic POS tagging. The majority of NLP applications require this process as a crucial step. This study intends to propose a machine learning-based Arabic POS tagger. YAMCHA tool is the machine learning system employed in this study. YAMCHA utilizes Support Vector Machines as a machine learning algorithm. SVM classifies data with high accuracy because it makes use of part of data in training process. As a result, in order to train the system, a substantial amount of annotated data must be evaluated at the POS level. A corpus of 100,039 words is utilized in this study. It was divided into training and testing parts, totaling 64,608 and 35,431 words, respectively. A tag set of 48 morphological tags were used in training and testing. To reach the best result in the automatic POS tagging, the system was trained multiple times with changing the range of linguistic information used in training process, and then new texts were tested and evaluated. The least error rate achieved was 11.4%. This rate was reached when the preceding word of the target one was considered in the training process without considering its POS tag (F: −1‥0: 0‥).","PeriodicalId":179850,"journal":{"name":"2022 20th International Conference on Language Engineering (ESOLEC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124616315","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-12DOI: 10.1109/ESOLEC54569.2022.10009462
Mahinaz Hegazy, S. Alansary
Due to the technological advancement and progress in NLP and text editing tools, there is an increasing demand for the paraphrasing practice. This demand has motivated researchers because numerous NLP applications are associated with it, including information retrieval, query answering, essay authenticity, text summarization, etc. This paper is a survey of several computational approaches for paraphrasing generation since the task of generating or identifying semantic equivalence for different linguistic elements is an essential part of NLP. It surveys a revised account including the most recent approaches to paraphrase generation up to Universal Networking Language Systems. The research specifically examines paraphrasing generation for Arabic using transformational rules. This is achieved by testing a paraphrasing on-line tool for Arabic and attempting an analysis and evaluation of its paraphrasing practice. The free online selected tool is Paraphrase Tool.com which is used to paraphrase more than one hundred languages, including Arabic. The resulting output is evaluated using BLEU to determine the accuracy of the rendered paraphrase and its semantic equivalence to the original source text.
{"title":"A Revised Survey of Paraphrasing Generation Approaches and Tools for Arabic","authors":"Mahinaz Hegazy, S. Alansary","doi":"10.1109/ESOLEC54569.2022.10009462","DOIUrl":"https://doi.org/10.1109/ESOLEC54569.2022.10009462","url":null,"abstract":"Due to the technological advancement and progress in NLP and text editing tools, there is an increasing demand for the paraphrasing practice. This demand has motivated researchers because numerous NLP applications are associated with it, including information retrieval, query answering, essay authenticity, text summarization, etc. This paper is a survey of several computational approaches for paraphrasing generation since the task of generating or identifying semantic equivalence for different linguistic elements is an essential part of NLP. It surveys a revised account including the most recent approaches to paraphrase generation up to Universal Networking Language Systems. The research specifically examines paraphrasing generation for Arabic using transformational rules. This is achieved by testing a paraphrasing on-line tool for Arabic and attempting an analysis and evaluation of its paraphrasing practice. The free online selected tool is Paraphrase Tool.com which is used to paraphrase more than one hundred languages, including Arabic. The resulting output is evaluated using BLEU to determine the accuracy of the rendered paraphrase and its semantic equivalence to the original source text.","PeriodicalId":179850,"journal":{"name":"2022 20th International Conference on Language Engineering (ESOLEC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115901173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-12DOI: 10.1109/ESOLEC54569.2022.10009633
Mohamed Fawzy, M. Fakhr, M. A. Rizka
Users share opinions and discussions on the internet through social media platforms. Nowadays, a significant number of internet users speak the Arabic language. They tend to express their opinions using different dialects. Therefore, understanding people's opinions and emotions become an urgent matter. The Arabic sentiment analysis is challenging because of linguistic complexity, data availability, and data quality, and it has multiple dialects. Therefore, research for low resources sentiment analysis became necessary. This study proposes a Bidirectional Encoder Representations from Transformers (BERT) that uses Convolutional Neural Network (CNN) as a classification head for Arabic low data resources for sentiment analysis. The classification head includes the CNN layer, drop-out layer, and a Relu activation function. The proposed approach experimented on three datasets collected from Twitter containing different dialects. The last four BERT layers were fined-tuned and while other layers were frozen. The suggested model outperforms current state-of-the-art models' accuracy with 50% fewer batch size, fewer training layers, and ∼20% fewer epochs.
{"title":"Sentiment Analysis For Arabic Low Resource Data Using BERT-CNN","authors":"Mohamed Fawzy, M. Fakhr, M. A. Rizka","doi":"10.1109/ESOLEC54569.2022.10009633","DOIUrl":"https://doi.org/10.1109/ESOLEC54569.2022.10009633","url":null,"abstract":"Users share opinions and discussions on the internet through social media platforms. Nowadays, a significant number of internet users speak the Arabic language. They tend to express their opinions using different dialects. Therefore, understanding people's opinions and emotions become an urgent matter. The Arabic sentiment analysis is challenging because of linguistic complexity, data availability, and data quality, and it has multiple dialects. Therefore, research for low resources sentiment analysis became necessary. This study proposes a Bidirectional Encoder Representations from Transformers (BERT) that uses Convolutional Neural Network (CNN) as a classification head for Arabic low data resources for sentiment analysis. The classification head includes the CNN layer, drop-out layer, and a Relu activation function. The proposed approach experimented on three datasets collected from Twitter containing different dialects. The last four BERT layers were fined-tuned and while other layers were frozen. The suggested model outperforms current state-of-the-art models' accuracy with 50% fewer batch size, fewer training layers, and ∼20% fewer epochs.","PeriodicalId":179850,"journal":{"name":"2022 20th International Conference on Language Engineering (ESOLEC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131198218","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-12DOI: 10.1109/ESOLEC54569.2022.10009530
Dalal Abdullah Aljohany, Hassanin M. Al-Barhamtoshy, Felwa A. Abukhodair
As Arabic is considered a low-resource and a rich morphology language. As result, Arabic is considered one of the most challenging languages in Machine Translation (MT). While numerous translation research concentrated on Indo-European languages, much less was made in Arabic. Therefore, the quality of Arabic Machine Translation (ArMT) continues to require improvement. Neural Machine Translation (NMT) is now the state-of-the-art in MT approaches. In this paper, we propose a model for two-way translation between the Arabic and English languages. The proposed model based on NMT and use the Long Short-Term Memory (LSTM) encoder-decoder model with attention mechanism. In the basic encoder–decoder performance is linked to the size of the input sentence, such that as the latter increases, performance diminishes swiftly. Attention mechanisms (AMs) are used to overcome this issue. The proposed model by combining LSTM and attention mechanism is capable to improve accuracy result of translation. The experimental results show that this proposed model improves accuracy of translation and reduces the loss.
{"title":"Arabic Machine Translation (ArMT) based on LSTM with Attention Mechanism Architecture","authors":"Dalal Abdullah Aljohany, Hassanin M. Al-Barhamtoshy, Felwa A. Abukhodair","doi":"10.1109/ESOLEC54569.2022.10009530","DOIUrl":"https://doi.org/10.1109/ESOLEC54569.2022.10009530","url":null,"abstract":"As Arabic is considered a low-resource and a rich morphology language. As result, Arabic is considered one of the most challenging languages in Machine Translation (MT). While numerous translation research concentrated on Indo-European languages, much less was made in Arabic. Therefore, the quality of Arabic Machine Translation (ArMT) continues to require improvement. Neural Machine Translation (NMT) is now the state-of-the-art in MT approaches. In this paper, we propose a model for two-way translation between the Arabic and English languages. The proposed model based on NMT and use the Long Short-Term Memory (LSTM) encoder-decoder model with attention mechanism. In the basic encoder–decoder performance is linked to the size of the input sentence, such that as the latter increases, performance diminishes swiftly. Attention mechanisms (AMs) are used to overcome this issue. The proposed model by combining LSTM and attention mechanism is capable to improve accuracy result of translation. The experimental results show that this proposed model improves accuracy of translation and reduces the loss.","PeriodicalId":179850,"journal":{"name":"2022 20th International Conference on Language Engineering (ESOLEC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121797645","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-12DOI: 10.1109/ESOLEC54569.2022.10009467
M. Taher, M. Hedaya, B. Bakeer, Passant El Kafrawy, Mahmoud Zakaria
In this work, the stochastic dispersion of novel coronavirus disease 2019 (COVID-19) at the borders between France and Italy has been considered using a multi-input multi-output stochastic model. The physical effects of wind, temperature and altitude have been investigated as these factors and physical relationships are stochastic in nature. Stochastic terms have also been included to take into account the turbulence effect, and the random nature of the above physical parameters considered. Then, a method is proposed to identify the developed model's order and parameters. The actual data has been used in the identification and prediction process as a reference. These data have been divided into two parts: the first part is used to calculate the stochastic parameters of the model which are used to predict the COVID-19 level, while the second part is used as a check data. The predicted results are in good agreement with the check data.
{"title":"Dynamic Modeling and Identification of the COVID-19 Stochastic Dispersion","authors":"M. Taher, M. Hedaya, B. Bakeer, Passant El Kafrawy, Mahmoud Zakaria","doi":"10.1109/ESOLEC54569.2022.10009467","DOIUrl":"https://doi.org/10.1109/ESOLEC54569.2022.10009467","url":null,"abstract":"In this work, the stochastic dispersion of novel coronavirus disease 2019 (COVID-19) at the borders between France and Italy has been considered using a multi-input multi-output stochastic model. The physical effects of wind, temperature and altitude have been investigated as these factors and physical relationships are stochastic in nature. Stochastic terms have also been included to take into account the turbulence effect, and the random nature of the above physical parameters considered. Then, a method is proposed to identify the developed model's order and parameters. The actual data has been used in the identification and prediction process as a reference. These data have been divided into two parts: the first part is used to calculate the stochastic parameters of the model which are used to predict the COVID-19 level, while the second part is used as a check data. The predicted results are in good agreement with the check data.","PeriodicalId":179850,"journal":{"name":"2022 20th International Conference on Language Engineering (ESOLEC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121957990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-12DOI: 10.1109/ESOLEC54569.2022.10009068
Fatma G. Rizk
Interdisciplinary studies have become a must; they are deep studies to link the exact specialization with other sciences so that the two fields will benefit the most. And here are the different types of corpora, one of the forms of inter-studies, through which the researcher tries to link between Greek and computer studies, and how they can be applied. Therefore, this research was employed to apply the use of the learner corpus as a work with a specific methodology planned with clear objectives in computer form as an application to university students. Why? To detect linguistic errors on the computer when translating Greek literary texts into Arabic, his results in the translation of their texts into a correct Arabic translation, in addition to defining this type of corpus and their most important characteristics and how much they benefit from Arab students studying the ancient Greek language. Indeed, this research came as an applied study on students of the sixth semester of Greek and Latin texts at the Faculty of Archeology - Fayoum University, where the student depends on the Greek source through (TLG) in addition to the use of the Perseus website, with the provision of grammatical books, dictionaries, to enable the teacher to identify the most important strengths and weaknesses that the student may face during the study, translate these texts, and store them inside the corpus. The study came as an application to the play Oedipus belonging to the Greek poet Sophocles.
{"title":"Learner Corpus in Teaching Greek to Arabic Natives: A Computational Linguistic Study of the Play Oedipus the King of Sophocles","authors":"Fatma G. Rizk","doi":"10.1109/ESOLEC54569.2022.10009068","DOIUrl":"https://doi.org/10.1109/ESOLEC54569.2022.10009068","url":null,"abstract":"Interdisciplinary studies have become a must; they are deep studies to link the exact specialization with other sciences so that the two fields will benefit the most. And here are the different types of corpora, one of the forms of inter-studies, through which the researcher tries to link between Greek and computer studies, and how they can be applied. Therefore, this research was employed to apply the use of the learner corpus as a work with a specific methodology planned with clear objectives in computer form as an application to university students. Why? To detect linguistic errors on the computer when translating Greek literary texts into Arabic, his results in the translation of their texts into a correct Arabic translation, in addition to defining this type of corpus and their most important characteristics and how much they benefit from Arab students studying the ancient Greek language. Indeed, this research came as an applied study on students of the sixth semester of Greek and Latin texts at the Faculty of Archeology - Fayoum University, where the student depends on the Greek source through (TLG) in addition to the use of the Perseus website, with the provision of grammatical books, dictionaries, to enable the teacher to identify the most important strengths and weaknesses that the student may face during the study, translate these texts, and store them inside the corpus. The study came as an application to the play Oedipus belonging to the Greek poet Sophocles.","PeriodicalId":179850,"journal":{"name":"2022 20th International Conference on Language Engineering (ESOLEC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126762993","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：