Arabic Language NLP

This study investigates the influence of various Natural Language Processing (NLP) models on the accuracy and efficiency of Arabic linguistic applications. Employing a systematic review and comparative analysis, the research evaluates rule-based, statistical, machine learning, deep learning, and large language models. Case studies across sentiment analysis, machine translation, text summarization, and named entity recognition highlight performance trade-offs. Key findings indicate that model selection critically depends on task-specific needs, data quality, and computational resources, with large language models offering notable accuracy gains but higher resource demands. The study emphasizes that balancing accuracy and efficiency is essential for practical deployment in Arabic NLP contexts. Its added value lies in providing a unified framework for assessing models, aiding developers and policymakers in optimizing system design for both performance and resource constraints.

This study investigates the multifaceted challenges of Arabic language processing in artificial intelligence (AI) systems, emphasizing linguistic, technical, and ethical dimensions. Employing a qualitative analysis of current research, it identifies key barriers such as morphological complexity, syntactic variation, dialect diversity, and limited annotated datasets. The research examines natural language processing tasks-tokenization, part-of-speech tagging, named entity recognition-alongside issues in machine translation, sentiment analysis, and speech recognition. Ethical considerations, including bias and privacy, are critically addressed. The findings underscore that overcoming these obstacles requires advancements in deep learning, collaborative cross-regional efforts, and the development of comprehensive linguistic resources. By mapping challenges to potential solutions, the study offers a roadmap for improving Arabic NLP performance, enabling equitable AI applications across Arabic-speaking communities.

is the first shared task that addresses the evaluation of plagiarism detection methods for Arabic texts. It has two sub- tasks, namely external plagiarism detection and intrinsic plagiarism detection. A total of 8 runs have been submitted and tested on the standardized corpora developed for the track. This overview paper describes these evaluation corpora, discusses the participants' methods, and highlights their building blocks that could be language dependent.

In the era of digital architecture, parametric design plays a fundamental role in the generative architectural design process. The most important of its benefits are that it allows a visual representation of the design process, a designer interaction during the construction of the parametric model using visual codes, the representation of steps sequentially and logically, in addition to the ability to adapt the model according to the client's requirements and the designer's wishes, and to generate various design solutions and alternatives for the same model. Parametric Typology is an approach that combines typological design with computational parametric techniques in order to produce various designs that belong to a specific building type. The research problem revolves around the application of parametric Typology to generat initial designs for local housing. Previous studies have proposed different approaches in the implementation of parametric typology to designs ranging from urban scale, individual buildings, to building elements. The research presented a theoretical framework, drawn from previous studies, to clarify the concepts and techniques of parametric typology method. The methodology was applied to build parametric model using a visual algorithm in grasshopper software. Prototypes for single-family housing had been generated as a functional type that occupies the largest space in cities; in addition there is a huge demand for it. As a result of the study, initial models were implemented with variety of parametric in terms of the placement of activities and the number of spaces.

As e-commerce is becoming more and more popular, the number of customer reviews that a product receives grows rapidly. In order to enhance customer satisfaction and their shopping experiences, it has become important to analysis customers reviews to extract opinions on the products that they buy. Thus, Opinion Mining is getting more important than before especially in doing analysis and forecasting about customers behavior for businesses purpose. The right decision in producing new products or services based on data about customers characteristics means profit for organization/company. This paper proposes a new architecture for Opinion Mining, which uses a multidimensional model to integrate customers characteristics and their comments about products (or services). The key step to achieve this objective is to transfer comments (opinions) to a fact table that includes several dimensions, such as, customers, products, time and locations. This research presents a comprehensive way to calculate customers orientation for all possible products attributes.

In the era of digital architecture, parametric design plays a fundamental role in the generative architectural design process. The most important of its benefits are that it allows a visual representation of the design process, a designer interaction during the construction of the parametric model using visual codes, the representation of steps sequentially and logically, in addition to the ability to adapt the model according to the client's requirements and the designer's wishes, and to generate various design solutions and alternatives for the same model. Parametric Typology is an approach that combines typological design with computational parametric techniques in order to produce various designs that belong to a specific building type. The research problem revolves around the application of parametric Typology to generat initial designs for local housing. Previous studies have proposed different approaches in the implementation of parametric typology to designs ranging from urban scale, individual buildings, to building elements. The research presented a theoretical framework, drawn from previous studies, to clarify the concepts and techniques of parametric typology method. The methodology was applied to build parametric model using a visual algorithm in grasshopper software. Prototypes for single-family housing had been generated as a functional type that occupies the largest space in cities; in addition there is a huge demand for it. As a result of the study, initial models were implemented with variety of parametric in terms of the placement of activities and the number of spaces.

UNPUBLISHED PAPER COMPARING THE HAND-ROLLED PARSER/POS TAGGER USED IN TUNISIYA.ORG WITH SOME ML METHODS

This paper presents a comparison of several different part-of-speech taggers trained on a hand-annotated Tunisian Arabic sample of 6,000 words in 450 sentences. Despite the small size of the annotated corpus, the Trigram Tagger and Brill Tagger performed nearly as well as a custom rule-based tagger. This work not only contributes a new resource to Tunisian Arabic-a severely under-resourced language-but also provides insight for developing materials for other under-resourced languages.

Measuring the amount of shared information between two documents is a key to address a number of Natural Language Processing (NLP) challenges such as Information Retrieval (IR), Semantic Textual Similarity (STS), Sentiment Analysis (SA) and Plagiarism Detection (PD). In this paper, we report a plagiarism detection system based on two layers of assessment: 1) Fingerprinting which simply compares the documents fingerprints to detect the verbatim reproduction; 2) Word embedding which uses the semantic and syntactic properties of words to detect much more complicated reproductions. Moreover, Word Alignment (WA), Inverse Document Frequency (IDF) and Part-of-Speech (POS) weighting are applied on the examined documents to support the identification of words that are most descriptive in each textual unit. In the present work, we focused on Arabic documents and we evaluated the performance of the system on a data-set of holding three types of plagiarism: 1) Simple reproduction (copy and past...

AraPlagDet is the first shared task that addresses the evaluation of plagiarism detection methods for Arabic texts. It has two subtasks, namely external plagiarism detection and intrinsic plagiarism detection. A total of 8 runs have been submitted and tested on the standardized corpora developed for the track. This overview paper describes these evaluation corpora, discusses the participants' methods, and highlights their building blocks that could be language dependent.

Easy access that Internet has provided to vast quantities of electronic data, textual plagiarism has become a major concern especially in academic documents and research and scientific institutions. So with increasing rate of amount of information on the Internet, the problems and disorders resulting from bilingual plagiarism have resulted in solutions that they can be detected with auto-detection methods. The recommended detection methods are mostly used for bilingual plagiarism in EnglishSpanish, Bengali, German, French, and Vietnamese etc. In this paper, a method has been proposed that based on overall reliance of textual contents provides and using vector space model (VSM) can automatically detect bilingual plagiarism English-Persian. Method after implementation assessed using test texts through accuracy and recalling standards and F-criterion reliability and the results showed that the proposed method can detect English Persian bilingual plagiarism with accuracy criteria of 0.8...

The paper investigates method for the style breach detection task. We developed a method based on mapping sentences into high dimensional vector space. Each sentence vector depends on the previous and next sentence vectors. As main architecture for this mapping we use the pre-trained encoder-decoder model. Then we use these vectors for constructing an author style function and detecting outliers. Method was tested on the PAN-2017 collection for the style breach detection task.

The anonymity of a text’s writer is an important topic for some domains, such as witness protection and anonymity programs. Stylometry can be used to reveal the true author of a text even if s/he wishes to hide his/her identity. In this paper, we present our approach for hiding an author’s identity by masking their style, which we developed for the Author Obfuscation task, part of the PAN-2016 competition. The approach consists of three main steps: the first one is an evaluation of different metrics in the text that can indicate authorship; the second one is application of various transformations, so that those metrics of the target text are adjusted towards the average level, while still keeping the meaning and the soundness of the text; as a final step, we are adding random noise to the text. Our system showed the best performance for masking the author style.

Various approaches have been implemented for plagiarism detection used, for author‘s work and academic publication, there is a purpose to create such reliable and performant plagiarism detection with increasing amount of publications. This is a serious offense where one author presents someone else’s work as his own. Moreover these algorithms don’t consider similar sections for efficient comparison. The proposed framework performs efficient sections wise plagiarism detection and provides suggestions for improving documents. The precision, recall and accuracy based on different n-gram features is presented showing the strictness of higher level n-gram features.

Opinion mining applications work with a large number of opinion holders. This means a summary of opinions is important so we can easily interpret holders' opinions. The aim of this paper is to provide a feature-based summarization for Arabic reviews. In our work, a system is proposed using Natural Language Processing (NLP) techniques, information extraction and sentiment lexicons. This provides users to access the opinions expressed in hundreds of reviews in a concise and useful manner. We start with extracting feature for a specific domain, assigned sentiment classification to each feature, and then summarized the reviews. We conducted a set of experiments to evaluate our system using data corpus from the hotel domain. The accuracy for opinion mining we calculated using objective evaluation was 71.22%. We, also, applied subjective evaluation for the summary generation and it indicated that our system achieved a relevant measure of 73.23% accuracy for positive summary and 72.46% accuracy for a negative summary.

In this paper, we have used one of the preconditioned conjugate gradient algorithm with the Quasi-Newton approximation; namely the BFGS preconditioned algorithm which was suggested by (AL-Bayati and Aref, 2001). In this paper we have suggested a new algorithm for constrained optimization with robust numerical results, for solving constrained optimization problems.

We describe our submitted algorithm to the text alignment sub-task of the plagiarism detection task in the PAN2014 challenge that achieved a plagdet score 0.855. By extracting contextual features for each document character and grouping those that are relevant for a given pair of documents, we generate seeds of atomic plagiarism cases. These are then merged by an agglomerative single- linkage strategy using a defined distance measure.

This paper presents our approach to the Author Clustering task at PAN 2017. We performed a hierarchical clustering analysis of different document features: typed and untyped character n-grams, and word n-grams. We experimented with two feature representation methods, log-entropy model, and tf-idf; while tuning minimum frequency threshold values to reduce the dimensionality. Our system was ranked 1 in both subtasks, author clustering and authorship-link ranking.

Text reuse is becoming a serious issue in many fields and research shows that it is much harder to detect when it occurs across languages. The recent rise in multilingual content on the Web has increased cross-language text reuse to an unprecedented scale. Although researchers have proposed methods to detect it, one major drawback is the unavailability of large-scale gold standard evaluation resources built on real cases. To overcome this problem, we propose a cross-language sentence/passage level text reuse corpus for the English-Urdu language pair. The Cross-Language English-Urdu Corpus (CLEU) has source text in English whereas the derived text is in Urdu. It contains in total 3,235 sentence/passage pairs manually tagged into three categories that is near copy, paraphrased copy, and independently written. Further, as a second contribution, we evaluate the Translation plus Mono-lingual Analysis method using three sets of experiments on the proposed dataset to highlight its usefulness. Evaluation results (f 1 =0.732 binary, f 1 =0.552 ternary classification) indicate that it is harder to detect cross-language real cases of text reuse, especially when the language pairs have unrelated scripts. The corpus is a useful benchmark resource for the future development and assessment of cross-language text reuse detection systems for the English-Urdu language pair.

PAN 2018 explores several authorship analysis tasks enabling a systematic comparison of competitive approaches and advancing research in digital text forensics. More specifically, this edition of PAN introduces a shared task in cross-domain authorship attribution, where texts of known and unknown authorship belong to distinct domains, and another task in style change detection that distinguishes between single-author and multi-author texts. In addition, a shared task in multimodal author profiling examines, for the first time, a combination of information from both texts and images posted by social media users to estimate their gender. Finally, the author obfuscation task studies how a text by a certain author can be paraphrased so that existing author identification tools are confused and cannot recognize the similarity with other texts of the same author. New corpora have been built to support these shared tasks. A relatively large number of software submissions (41 in total) was received and evaluated. Best paradigms are highlighted while baselines indicate the pros and cons of submitted approaches. (stylistic fingerprint) but she also shares some properties with other people of similar background (age, gender, education, etc.) It is quite challenging to define or measure both personal style (for each individual author) and collective style (males, females, young people, old people, etc.). In addition, it remains unclear what one should modify in her texts in order to attempt to hide her identity or to mimic the style of another author. This edition of PAN deals with these challenging issues. Author identification puts emphasis on the personal style of individual authors. The most common task is authorship attribution where there is a set of candidate authors (suspects), with samples of their texts, and one of them is selected as the most likely author of a text of disputed authorship [31]. This can be a closed-set (one of the suspects is surely the true author) or an open-set (the true author may not be among the suspects) attribution case. This edition of PAN focuses on closed-set cross-domain authorship attribution, that is, when the texts unquestionably written by the suspects and the texts of disputed authorship belong to different domains. This is a realistic scenario suitable for several applications. For example, imagine the case of a crime novel published anonymously when all candidate authors have only published fantasy novels [13] or a disputed tweet when the available texts written by the suspects are newspaper articles. To be able to control the domain of texts, we turned to so-called fanfiction [11]. This term refers to the large body of contemporary fiction that is nowadays created by non-professional authors ('fans'), who write in the tradition of a well-known source work, such as the Harry Potter series by J.K. Rowling, that is sometimes called the 'canon'. These writings or 'fics' within such a 'fandom' heavily borrow characters, motives, settings, etc. from the source fandom. Fanfiction provides excellent material to study cross-domain authorship attribution since most fans are active in multiple fandoms. Another important dimension in author identification is to intrinsically analyse a document, possibly written by multiple authors and identify the contribution of each co-author. The previous edition of PAN aimed to find the exact border positions within a document where the authorship changes. Taking the respective results into account which have shown that the problem is quite hard [39], we substantially relaxed the task this year and broke it down to the simple question: Given a document, are there any style changes or not? An alternative formulation would thus be to solely predict whether a document is written by a single author or by multiple collaborators, whereby it is irrelevant to the task to identify the exact border positions between authors. While the evaluation of the two preceding tasks relied on the Webis-TRC-12 data set [21], we created a novel data set by utilizing the StackExchange network 1. Containing millions of publicly available questions and answers regarding several topics and subtopics, it represents a rich source which we exploited to build a comprehensive, but still realistic data set for the style change detection task. When the collective style of groups of authors is considered, author profiling attempts to predict demographic and social characteristics, like age, gender, education, and personality traits. It is a research area associated with important applications in social media analytics and marketing as well as cyber forensics. In this edition of PAN, for the first time, multimodal information is considered. Both texts and images posted by social media users are used to predict their gender.

Plagiarism Detection Systems are critical in identifying instances of plagiarism, particularly in the educational sector whenever it comes to scientific publications and papers. Plagiarism occurs when any material is copied without the author's consent or attribution. To identify such acts, thorough knowledge of plagiarism types and classes is required. It is feasible to detect several sorts of plagiarism using current tools and methodologies. With the advancement of information and communication technologies (ICT) and the availability of online scientific publications, access to these publications has grown more convenient. Additionally, with the availability of several software text editors, plagiarism detection has become a crucial concern. Numerous scholarly articles have previously examined plagiarism detection and the two most often used datasets for plagiarism detection, WordNet and the PAN Dataset. The researchers described verbatim plagiarism detection as a straightforward case of copying and pasting, and then shed light on clever plagiarism, which is more difficult to detect since it may involve original text alteration, borrowing ideas from other studies, and Other scholars have said that plagiarism can obscure the scientific content by substituting terms, deleting or introducing material, rearranging or changing the original publications. The suggested system incorporated natural language processing (NLP) and machine learning (ML) techniques, as well as an external plagiarism detection strategy based on text mining and similarity analysis. The suggested technique employs a mix of Jaccard and cosine similarity. It was examined using the PAN-PC-11 corpus. The proposed system outperforms previous systems on the PAN-PC-11, as demonstrated by the findings. Additionally, the proposed system obtains an accuracy of 0.96, a recall of 0.86, an F-measure of 0.86, and a PlagDet score of 0.86. (0.86). 0.865 and the proposed technique is substantiated by a design application that is used to detect plagiarism in scientific publications and generate nonmedication notifications. Portable Document Format (PDF) .

Many approaches are characteristic of name opinion is based only on the review of the single-shaft, ignoring non-trivial disparities in the distribution of the word of those around Corpus different. In Proposed work a new technique introduced to determine the characteristics of the idea of the magazine online by using the difference in those statistics through two and more than two different entities, a corpus of specific domain entity and a free domain of the corpus contrasted. Then determine the inconsistency through a measurement called relevance domain (DR), which characterizes the relevance of the term for a collection of manuscripts. Compile a list of candidates for the review of the terms of the domain corpus review of a set of rules of syntax dependency. For each function extracted candidates, Then evaluate the intrinsic domain and extrinsic domain relevance, in the entities of domain-dependent and independent respectively. The candidates which are different and more specific to the domain are confirmed as the hallmark of those.

This thesis deals with two major topics: plagiarism detection in Arabic documents, and plagiarism detection based on the writing style changes in the suspicious document, which is called intrinsic plagiarism detection. This approach is an alternative to the text-matching approach, notably, in the absence of the plagiarism source. Our key contributions in these two areas lie first, in the development of Arabic corpora to allow for the evaluation of plagiarism detection software on this language and, second, in the development of a language-independent intrinsic plagiarism detection method that exploits the character n-grams in a machine learning approach while avoiding the curse of dimensionality. Our third key contribution is an investigation on which character n-grams, in terms of their frequency and length, are the best to detect plagiarism intrinsically. We carried out our experiments on standardised English corpora and also on the developed Arabic corpora using the method we developed and one of the most prominent intrinsic plagiarism detection methods. The findings of our analysis can be exploited by the future intrinsic plagiarism detection methods that use character n-grams. In addition to the above-mentioned technical contributions, we provide the reader with comprehensive and critical surveys of the literature of Arabic plagiarism detection and intrinsic plagiarism detection, which were lacking in both topics.

المحاور:

المحور الأول: - بعض المفاهيم المتعلقة بالبرمجة الآلية للغة.
المحور الثاني: - المعالجة الآلية لمنظومة الصرف.
المحور الثالث: - المعالجة الآلية لمنظومة النحو.
المحور الرابع: - المعالجة الآلية لمنظومة الكتابة.
المحور الخامس : - ميكنة المعجم.
المحور السادس : - الترجمة الآلية.

Extreme Learning Machine (ELM) is a new learning algorithm for feed forward neural network for classification or regression with a single layer of hidden nodes where the weights connecting inputs to hidden nodes are randomly assigned. Extreme Learning Machine allow polarity associated with natural language concept to be calculated in a more dynamic way to explore twitter user classification using content and context which are induced by hash tags and social communication in twitter. However, the explosive nature of the unstructured information generated around the world brings a challenge for sentiment-based information retrieval and analysis. This aims to solve the problem of extraction from product reviews that exploits common-sense knowledge by enabling fast reconfiguration of vector space.

The Web considers one of the main sources of customer opinions and reviews which they are represented in two formats; structured data (numeric ratings) and unstructured data (textual comments). Millions of textual comments about goods and services are posted on the web by customers and every day thousands are added, make it a big challenge to read and understand them to make them a useful structured data for customers and decision makers. Sentiment analysis or Opinion mining is a popular technique for summarizing and analyzing those opinions and reviews. In this paper, we use natural language processing techniques to generate some rules to help us understand customer opinions and reviews (textual comments) written in the Arabic language for the purpose of understanding each one of them and then convert them to a structured data. We use adjectives as a key point to highlight important information in the text then we work around them to tag attributes that describe the subject of the reviews, and we associate them with their values (adjectives).

Opinion target is defined as the object about which user expresses their opinions, typically as nouns or noun phrases. Opinion words are the words that are used to express user's opinions. Constructing an opinion words lexicon is also important because the lexicon is beneficial for identifying opinion expressions. Opinion mining is also called sentiment analysis, involves building a system to collect and categorize opinions about a product. Existing system uses Sentence-level extraction and corpus –level extraction. Sentence level extraction is the task of opinion target/word extraction is to identify the opinion target mentions or opinion expressions in sentences. These tasks are usually regarded as sequence-labeling problems. Sequence labeling models are used to build the extractor, such HMM. Corpus-level extraction did not identify the opinion target/word mentions in sentences, but aimed to extract a list of opinion targets or generate a sentiment word lexicon from texts. To overcome these problems by using alignment models.

Mining patterns are the main source of opinion feature extraction techniques, which was individually evaluated corpus mostly belong to evaluated corpus. A measure called Domain Relevance is used to identify candidate features from domain dependent and domain independent corpora both. Opinion Features originated are relevant to a domain. For every extracted candidate feature its individual Intrinsic Domain Relevance and Extrinsic Domain Relevance values are registered. Threshold has been compared with these values and recognizes as best candidate features. In this thesis, By applying feature filter creation the features from online reviews can be identified .

AraPlagDet is the first shared task that addresses the evaluation of plagiarism detection methods for Arabic texts. It has two subtasks, namely external plagiarism detection and intrinsic
plagiarism detection. A total of 8 runs have been submitted and
tested on the standardized corpora developed for the track. This overview paper describes these evaluation corpora, discusses the participants’ methods, and highlights their building blocks that could be language dependent.

‫ملخص‬ ‫الورقة‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫اﻟﻠﻐﺔ‬ ‫ﳌﻌﺎﳉﺔ‬ ‫ﺣﺎﺳﻮﺑﻴﺔ‬ ‫ﺑﺄﲝﺎث‬ ‫ﻟﻴﺪز‬ ‫ﲜﺎﻣﻌﺔ‬ ‫اﳊﺎﺳﻮب‬ ‫ﻟﻜﻠﻴﺔ‬ ‫اﻟﺘﺎﺑﻊ‬ ‫اﻟﻠﻐﺔ‬ ‫أﲝﺎث‬ ‫ﻳﻖ‬ ‫ﻓﺮ‬ ‫أﻋﻀﺎء‬ ‫ﻳﻬﺘﻢ‬ . ‫اﳌﺎﺿﻲ‬ ‫ﰲ‬ ‫ﻗﻤﻨﺎ‬ ‫ﻓﻌﻨﺪﻣﺎ‬ ‫ﻛﻨ‬ ‫أدر‬ ‫ﻟﻜﻨﻨﺎ‬ ‫و‬ ‫ات،‬ ‫اﻷدو‬ ‫ﻫﺬﻩ‬ ‫ﻣﻦ‬ ً ‫ﺟﺪا‬ ً ‫ﻗﻠﻴﻼ‬ ً ‫ﻋﺪدا‬ ‫وﺟﺪﻧﺎ‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫ﻟﻠﻐﺔ‬ ‫اﻟﻨﺼﻴﺔ‬ ‫ﻟﻠﻤﻌﺎﳉﺔ‬ ً ‫ﳎﺎﻧﺎ‬ ‫اﳌﺘﺎﺣﺔ‬ ‫ات‬ ‫ﻟﻸدو‬ ‫اﺳﻌﺔ‬ ‫و‬ ‫اﺳﺔ‬ ‫در‬ ‫ﺑﻌﻤﻞ‬ ‫أﻧﻪ‬ ‫ﺎ‬ ‫اﻵﱄ‬ ‫اﻟﺘﻌﻠﻢ‬ ‫ﺑﺮﳎﻴﺎت‬ ‫ﺗﻄﻮﻳﻊ‬ ‫ﳝﻜﻦ‬ ) machine learning ( ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫اﻟﻠﻐﺔ‬ ‫ﻋﻠﻰ‬ ‫ذﻟﻚ‬ ‫وﺗﻄﺒﻴﻖ‬ ‫ﻋﺎم‬ ‫ﺑﺸﻜﻞ‬ ‫اﻟﻠﻐﻮﻳﺔ‬ ‫ﻟﻠﻠﻤﻌﺎﳉﺔ‬ . ‫ﺑﻴﺔ‬ ‫ﻋﺮ‬ ‫ﻧﺼﻮص‬ ‫ﳎﻤﻮﻋﺔ‬ ‫أول‬ ‫ﲜﻤﻊ‬ ‫ﻗﻤﻨﺎ‬ ‫ﰒ‬ ‫وﻣﻦ‬ ‫اﻟﱪﳎﻴﺎت،‬ ‫ﻫﺬﻩ‬ ‫ﻳﺐ‬ ‫ﻟﺘﺪر‬ ‫ﻋﺮﰊ‬ ‫ﻧﺺ‬ ‫وﺟﻮد‬ ‫ﻳﺴﺘﺪﻋﻲ‬ ‫اﻷﻣﺮ‬ ‫وﻫﺬا‬ ) corpus ( ‫اﺠﻤﻟ‬ ‫ﻟﻠﺘﺤﻤﻴﻞ‬ ‫وﻣﺘﺎﺣﺔ‬ ‫اﳌﺼﺪر‬ ‫ﻣﻔﺘﻮﺣﺔ‬ ‫وﺟﻌﻠﻨﺎﻫﺎ‬ ‫ﻣﻨﺴﺠﻢ‬ ‫ﺑﺸﻜﻞ‬ ‫اﻟﻨﺺ‬ ‫ﻟﻌﺮض‬ ‫ﻣﺮﳛﺔ‬ ‫اﺟﻬﺔ‬ ‫و‬ ‫ﻧﺎﻣﺞ‬ ‫ﺑﺮ‬ ‫ﻧﺎ‬ ‫ﻃﻮر‬ ‫وﻛﺬﻟﻚ‬ ‫ﺎﱐ‬ . ‫اﺠﻤﻟﻤﻮﻋﺔ‬ ‫ﻫﺬﻩ‬ ‫اﻟﻨﺘﺎﺋﺞ‬ ‫وﺗﻘﻴﻴﻢ‬ ‫أﺧﺮى‬ ‫ﺣﺎﺳﻮﺑﻴﺔ‬ ‫اﻣﺞ‬ ‫ﺑﺮ‬ ‫ﻳﺐ‬ ‫ﻟﺘﺪر‬ ‫اﺳﺘﺨﺪﻣﻮﻫﺎ‬ ‫و‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫ﻟﻠﻐﺔ‬ ‫اﳊﺎﺳﻮﺑﻴﺔ‬ ‫اﳌﻌﺎﳉﺔ‬ ‫اء‬ ‫ﺧﱪ‬ ‫ﻗﺒﻞ‬ ‫ﻣﻦ‬ ‫اﺳﻌﺔ‬ ‫و‬ ‫ﺷﻬﺮة‬ ‫اﻛﺘﺴﺒﺖ‬ . ‫اﻟﺼ‬ ‫اﻟﺘﺤﻠﻴﻞ‬ ‫ﻣﺜﻞ‬ ‫اﳌﻌﺎﺻﺮ‬ ‫اﻟﻌﺮﰊ‬ ‫اﻟﻨﺺ‬ ‫ﻟﺘﺤﻠﻴﻞ‬ ‫ات‬ ‫أدو‬ ‫ﺑﺘﻄﻮﻳﺮ‬ ‫ﻗﻤﻨﺎ‬ ‫أﻧﻨﺎ‬ ‫ﻛﻤﺎ‬ ‫اﺳﻌﺔ‬ ‫اﻟﻮ‬ ‫ﻟﻠﺘﻐﻄﻴﺔ‬ ‫ﻟﻐﻮﻳﺔ‬ ‫وذﺧﺎﺋﺮ‬ ‫اﻟﻌﻨﻮﻧﺔ‬ ‫و‬ ‫اﻟﺘﺠﺬﻳﺮ‬ ‫ﺮﰲ،‬ ‫اﳋﻄﺎﺑﻴﺔ‬ ‫ﺑﺎﻟﻌﻼﻗﺎت‬ ‫ﻣﻌﻨﻮﻧﺔ‬ ‫ﻧﺼﻴﺔ‬ ‫ﳎﻤﻮﻋﺔ‬ ‫ﺗﻄﻮﻳﺮ‬ ‫و‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫ﻟﻠﻐﺔ‬ ) discourse relations ( ‫اﻟﻌﻼﻗﺎت‬ ‫ار‬ ‫ﻏﺮ‬ ‫ﻋﻠﻰ‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫ﻟﻠﻐﺔ‬ ‫ﺑﻨﺴﻠﻔﺎﻧﻴﺎ‬ ‫ﺟﺎﻣﻌﺔ‬ ‫ﻣﻦ‬ ‫اﻟﺼﺎدرة‬ ‫ﻳﺔ‬ ‫اﻹﳒﻠﻴﺰ‬ ‫ﻟﻠﻐﺔ‬ ‫اﳋﻄﺎﺑﻴﺔ‬ . ‫ﻧﻌﺘﺰ‬ ‫اﻟﱵ‬ ‫اﺋﺪة‬ ‫اﻟﺮ‬ ‫اﻟﺒﺤﺜﻴﺔ‬ ‫اﺠﻤﻟﺎﻻت‬ ‫وﻣﻦ‬ ‫ﻟﺘﻘﻨﻴﺔ‬ ‫إﻣﺘﺪاد‬ ‫وﺗﻌﺘﱪ‬ ‫اﻟﻜﺮﱘ‬ ‫ﻟﻠﻘﺮآن‬ ‫اﻟﻠﻐﻮﻳﺔ‬ ‫ﻟﻠﻤﻌﺎﳉﺔ‬ ‫ﺣﺎﺳﻮﺑﻴﺔ‬ ‫ﻳﻊ‬ ‫ﻣﺸﺎر‬ ‫ﻫﻮ‬ ‫ﻟﻴﺪز‬ ‫ﺟﺎﻣﻌﺔ‬ ‫ﰲ‬ ‫ﺎ‬ ‫اﻟﺘﻘﻠﻴﺪﻳﺔ‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫ﻟﻠﻐﺔ‬ ‫اﻟﻨﺼﻲ‬ ‫اﻟﺘﻨﻘﻴﺐ‬ . ‫ﻣﻮﻗﻊ‬ ‫و‬ ‫آﻟﻴﺔ‬ ‫ﳏﺎورة‬ ‫ﻧﺎﻣﺞ‬ ‫ﺑﺮ‬ ‫ﺗﻄﻮﻳﺮ‬ ‫اﻷﲝﺎث‬ ‫ﻫﺬﻩ‬ ‫وﻣﻦ‬ " ‫ﻗﺮآﱐ‬ " ‫ﻋﻠﻰ‬ ‫اﻟﺒﺤﺚ‬ ‫ﻳﺘﻴﺢ‬ ‫اﻟﺬي‬ ‫و‬ ‫اﻟ‬ ‫اﳌﻌﺮﻓﺔ‬ ‫ﻟﺘﻤﺜﻴﻞ‬ ‫وإﻃﺎر‬ ، ً ‫ﻣﺴﺒﻘﺎ‬ ‫ﻣﻌﺪة‬ ‫ﻣﻔﺎﻫﻴﻢ‬ ‫ﻣﺴﺘﻮى‬ ‫اﻟﻨﺤﻮﻳﺔ‬ ‫اﻟﻌﻨﻮﻧﺔ‬ ‫و‬ ‫ﻘﺮآﻧﻴﺔ‬ . ‫ﻣﻮﻗﻊ‬ ‫ﺑﺘﺪﺷﲔ‬ ‫ﻗﻤﻨﺎ‬ ً ‫ا‬ ‫وﻣﺆﺧﺮ‬ " ‫ﻟﻠﻘﺮآن‬ ‫ﺑﻴﺔ‬ ‫ﻋﺮ‬ ‫ﻟﻐﻮﻳﺔ‬ ‫ذﺧﲑة‬ ‫اﻟﻜﺮﱘ‬ ) " Quranic Arabic Corpus ] ( http //: corpus . quran . com [ ‫ﻟﻠﺘﺤﻤﻴﻞ‬ ‫ﻗﺎﺑﻞ‬ ‫إﻟﻜﱰوﱐ‬ ‫ﻣﻮرد‬ ‫وﻳﻌﺘﱪ‬ ‫ﻗﺮآﻧﻴﺔ‬ ‫ﻛﻠﻤﺔ‬ ‫ﻟﻜﻞ‬ ‫اﻟﻜﻼم‬ ‫أﻗﺴﺎم‬ ‫و‬ ‫اﻟﺼﺮف‬ ‫ﻣﺴﺘﻮى‬ ‫ﻋﻠﻰ‬ ‫ﺗﻔﺼﻴﻠﻴﺔ‬ ‫ﻟﻐﻮﻳﺔ‬ ‫ﲟﻌﻠﻮﻣﺎت‬ ‫وﻣﻮﺳﻢ‬ ‫اﺠﻤﻟﺎﱐ‬ . ‫وﻣﻨ‬ ‫ﺑﻘﺒﻮل‬ ‫ﺣﻈﻲ‬ ‫اﳌﻮﻗﻊ‬ ‫اﻧﻄﻼق‬ ‫ﺬ‬ ‫وﻫﻮ‬ ‫ﺑﻴﺔ،‬ ‫ﺑﺎﻟﻌﺮ‬ ‫اﻟﻨﺎﻃﻘﲔ‬ ‫ﻏﲑ‬ ‫ﻣﻦ‬ ‫اﻟﻘﺮآن‬ ‫ﻟﻐﺔ‬ ‫ﺗﻌﻠﻢ‬ ‫ﰲ‬ ‫اﻏﺒﲔ‬ ‫اﻟﺮ‬ ‫وﻛﺬﻟﻚ‬ ‫اﻟﻠﺴﺎﻧﻴﺎت‬ ‫أﲝﺎث‬ ‫ﰲ‬ ‫اء‬ ‫ﺧﱪ‬ ‫ﻣﻦ‬ ً ‫اﺳﻌﺎ‬ ‫و‬ ً ‫ا‬ ‫ﲨﻬﻮر‬ ‫اﺳﺘﻘﻄﺐ‬ ‫و‬ ‫اﺳﻊ‬ ‫و‬ ‫ﻣﻦ‬ ‫أﻛﺜﺮ‬ ‫وﳚﻠﺐ‬ ‫ﻟﻠﺒﺤﺚ‬ ‫ﺟﻮﺟﻞ‬ ‫ﳏﺮك‬ ‫ﻋﻠﻰ‬ ‫اﻟﻘﺮآن‬ ‫اﻋﺪ‬ ‫ﻗﻮ‬ ‫اﺳﺔ‬ ‫ﻟﺪر‬ ‫اﻷول‬ ‫اﳌﻮﻗﻊ‬ 50 ً ‫ﻳﺎ‬ ‫ﺷﻬﺮ‬ ‫اﺋﺮ‬ ‫ز‬ ‫أﻟﻒ‬ . ‫ﻧﻘﱰح‬ ‫ﺟﻌﻠﻨﺎ‬ ‫وﻫﺬا‬ " ‫ﻓﻬﻢ‬ ‫اﻟﻘﺮآن‬ " ‫ﻟﻌﺎم‬ ‫اﻟﺼﻨﺎﻋﻲ‬ ‫اﻟﺬﻛﺎء‬ ‫و‬ ‫اﳊﺎﺳﻮب‬ ‫ﻟﻌﻠﻮم‬ ‫ﻛﺒﲑ‬ ‫ﻛﺘﺤﺪ‬ 2010 ‫ﺑﻌﺪﻩ‬ ‫وﻣﺎ‬ . ‫أﻧﻨﺎ‬ ‫ﺣﻴﺚ‬ ‫اﳌﺸﺮوع‬ ‫ﻫﺬا‬ ‫ﺑﺘﻮﺳﻴﻊ‬ ‫ﻧﻘﻮم‬ ً ‫وﺗﺪرﳚﻴﺎ‬ ‫اﻟﻘﺮآﻧﻴﺔ‬ ‫اﳉﻤﻠﺔ‬ ‫ﻟﺒﻨﻴﺔ‬ ‫ﺗﺒﻌﻴﺔ‬ ‫ﺑﻴﺎﻧﻴﺔ‬ ‫ات‬ ‫ﺷﺠﺮ‬ ‫إﻋﺪاد‬ ‫ﻃﻮر‬ ‫ﰲ‬ ‫اﻵن‬ )) Quranic Arabic Dependency Treebank ‫ﻟﻠﻘﺮآء‬ ‫وﳑﻜﻨﺔ‬ ‫اﻟﻜﺮﱘ‬ ‫اﻟﻘﺮآن‬ ‫اب‬ ‫إﻋﺮ‬ ‫ﻛﺘﺐ‬ ‫ﻣﻦ‬ ‫ﻣﺴﺘﻤﺪة‬ ‫اﻵﻟﻴﺔ‬ ‫ة‬ . ‫ﺑﲔ‬ ‫اﺑﻂ‬ ‫رو‬ ‫ﺷﺒﻜﺔ‬ ‫وﺑﻨﺎء‬ ‫ﻗﺮآﻧﻴﺔ‬ ‫ﻣﻔﺎﻫﻴﻢ‬ ‫ﺑﺎﺳﺘﺨﻼص‬ ‫ﻧﻘﻮم‬ ‫أﻧﻨﺎ‬ ‫ﻛﻤﺎ‬ ‫اﳌﻔﺎﻫﻴﻢ‬ ‫ﳍﺬﻩ‬ ‫اﻟﻘﺮآن‬ ‫ﺿﻤﺎﺋﺮ‬ ‫وﻋﻮدة‬ ‫اﳌﻔﺎﻫﻴﻢ‬ ‫ﻫﺬﻩ‬ . ‫اﻟﻘﺮآن‬ ‫ﳌﻔﺮدات‬ ‫ﻟﻐﻮﻳﺔ‬ ‫ذﺧﲑة‬ ‫ﺑﻨﺎء‬ ‫اﻟﻘﺮآﱐ‬ ‫اﳌﺸﺮوع‬ ‫ﻫﺬا‬ ‫ﰲ‬ ‫اﳌﺴﺘﻘﺒﻠﻴﺔ‬ ‫اﳋﻄﻂ‬ ‫وﻣﻦ‬ ‫ﻣﺸﺮوع‬ ‫ار‬ ‫ﻏﺮ‬ ‫ﻋﻠﻰ‬ WordNet ‫ﳌﻔﺮدا‬ ‫دﻻﻟﻴﺔ‬ ‫ات‬ ‫إﻃﺎر‬ ‫ذﺧﲑة‬ ‫ﺗﻄﻮﻳﺮ‬ ‫وﻛﺬﻟﻚ‬ ‫ﻣﺸﺮوع‬ ‫ار‬ ‫ﻏﺮ‬ ‫ﻋﻠﻰ‬ ‫اﻟﻘﺮآن‬ ‫ت‬ FrameNet ‫ﻋﻠﻰ‬ ‫اﻟﻘﺮآﱐ‬ ‫اﻟﺒﺤﺚ‬ ‫ﻳﺘﻴﺢ‬ ‫ﻣﺘﻜﺎﻣﻞ‬ ‫ﲝﺚ‬ ‫ﳏﺮك‬ ‫ﺗﻄﻮﻳﺮ‬ ‫ﰒ‬ ‫وﻣﻦ‬ ‫اﳋﻄﺎﺑﻴﺔ،‬ ‫ﺑﺎﻟﻌﻼﻗﺎت‬ ‫ﻣﻌﻨﻮﻧﺔ‬ ‫ﻟﻠﻘﺮآن‬ ‫ﻧﺼﻴﺔ‬ ‫ﳎﻤﻮﻋﺔ‬ ‫ﺗﻄﻮﻳﺮ‬ ‫وﻛﺬﻟﻚ‬ ‫اﻟﺬﻛﺮ‬ ‫آﻧﻔﺔ‬ ‫اﻟﻠﻐﻮﻳﺔ‬ ‫اﻟﺬﺧﺎﺋﺮ‬ ‫ﺧﻼل‬ ‫ﻣﻦ‬ ‫وذﻟﻚ‬ ‫اﳌﻔﺎﻫﻴﻢ‬ ‫و‬ ‫اﻟﺼﺮف‬ ‫و‬ ‫اﻟﻨﺤﻮ‬ ‫و‬ ‫اﻷﺳﻠﻮب‬ ‫و‬ ‫اﳌﻔﺮدات‬ ‫ﻣﺴﺘﻮي‬ . ‫ﻧﺎﺣﺠﺔ‬ ‫ﻳﻊ‬ ‫ﻣﺸﺎر‬ ‫ﺧﻼل‬ ‫وﻣﻦ‬ ‫اﻟﺘﻘﻠﻴﺪﻳﺔ‬ ‫ﻟﻠﻨﺼﻮص‬ ‫اﳊﺎﺳﻮﺑﻴﺔ‬ ‫اﳌﻌﺎﳉﺔ‬ ‫ﰲ‬ ‫اﳌﻜﺘﺴﺒﺔ‬ ‫اﳋﱪة‬ ‫ﺗﻄﺒﻴﻖ‬ ‫ﻧﻄﻤﺢ‬ ‫اﻟﻜﺮﱘ‬ ‫اﻟﻘﺮآن‬ ‫ﻋﻠﻰ‬ -‫اث‬ ‫اﻟﱰ‬ ‫وﻛﺘﺐ‬ ‫اﻟﻨﺒﻮي‬ ‫اﳊﺪﻳﺚ‬ ‫ﻣﺜﻞ‬ ‫اﻟﻌﺮﰊ‬ -‫اﳌﻌﺎﺻﺮة‬ ‫اﻟﻔﺼﺤﻰ‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫وﻛﺬﻟﻚ‬ . ‫اﻟﺘﻨﻘﻴﺐ‬ ‫و‬ ‫اﻟﺒﺤﺚ‬ ‫اض‬ ‫أﻏﺮ‬ ‫ﺗﻠﱯ‬ ‫ﻣﻌﻨﻮﻧﺔ‬ ‫ﻧﺼﻴﺔ‬ ‫ﳎﻤﻮﻋﺔ‬ ‫ﻟﺘﻄﻮﻳﺮ‬ ‫ﳔﻄﻂ‬ ‫اﳌﺜﺎل‬ ‫ﺳﺒﻴﻞ‬ ‫ﻓﻌﻠﻰ‬ ‫اﻹ‬ ‫ﳎﺎل‬ ‫ﰲ‬ ‫ﻟﻠﻤﺨﺘﺼﲔ‬ ‫اﻟﻨﺼﻲ‬ ‫اﻹﺧﺘﺼﺎص‬ ‫ﻫﺬا‬ ‫ﰲ‬ ‫اﳌﻌﻨﻮﻧﺔ‬ ‫ﻟﻠﻤﻔﺎﻫﻴﻢ‬ ‫اﳌﻌﺮﰲ‬ ‫اﻟﺘﻤﺜﻴﻞ‬ ‫ﺧﻼل‬ ‫ﻣﻦ‬ ‫وذﻟﻚ‬ ‫اﻹﺳﻼﻣﻲ‬ ‫ﻗﺘﺼﺎد‬ . ‫وﻛﺬﻟﻚ‬ ‫ﺗﻘﻨﻴﺎت‬ ‫ﺗﻮﻇﻴﻒ‬ ‫ﻣﻦ‬ ‫ﳝﻜﻨﻨﺎ‬ ‫وﻫﺬا‬ ‫اﻹﳒﻴﻞ‬ ‫و‬ ‫اة‬ ‫اﻟﺘﻮر‬ ‫ﻣﺜﻞ‬ ‫اﻷﺧﺮى‬ ‫اﻷدﻳﺎن‬ ‫ﰲ‬ ‫ﻣﻘﺪﺳﺔ‬ ‫ﻛﺘﺐ‬ ‫ﻟﺘﺸﻤﻞ‬ ‫اﳊﺎﺳﻮﺑﻴﺔ‬ ‫أﲝﺎﺛﻨﺎ‬ ‫ﲤﺪﻳﺪ‬ ‫ﻋﻠﻴﻨﺎ‬ ‫ﺮض‬ ُ ‫ﻋ‬ ‫اﳌﺼﺎدر‬ ‫ﻫﺬﻩ‬ ‫ﺑﲔ‬ ‫اﻟﻔﺮوﻗﺎت‬ ‫و‬ ‫اﻟﺘﺸﺎﺑﻪ‬ ‫ﳌﻼﺣﻈﺔ‬ ‫ﺣﺎﺳﻮﺑﻴﺔ‬ .

‫ملخص‬ ‫الورقة‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫اﻟﻠﻐﺔ‬ ‫ﳌﻌﺎﳉﺔ‬ ‫ﺣﺎﺳﻮﺑﻴﺔ‬ ‫ﺑﺄﲝﺎث‬ ‫ﻟﻴﺪز‬ ‫ﲜﺎﻣﻌﺔ‬ ‫اﳊﺎﺳﻮب‬ ‫ﻟﻜﻠﻴﺔ‬ ‫اﻟﺘﺎﺑﻊ‬ ‫اﻟﻠﻐﺔ‬ ‫أﲝﺎث‬ ‫ﻳﻖ‬ ‫ﻓﺮ‬ ‫أﻋﻀﺎء‬ ‫ﻳﻬﺘﻢ‬ . ‫اﳌﺎﺿﻲ‬ ‫ﰲ‬ ‫ﻗﻤﻨﺎ‬ ‫ﻓﻌﻨﺪﻣﺎ‬ ‫ﻛﻨ‬ ‫أدر‬ ‫ﻟﻜﻨﻨﺎ‬ ‫و‬ ‫ات،‬ ‫اﻷدو‬ ‫ﻫﺬﻩ‬ ‫ﻣﻦ‬ ً ‫ﺟﺪا‬ ً ‫ﻗﻠﻴﻼ‬ ً ‫ﻋﺪدا‬ ‫وﺟﺪﻧﺎ‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫ﻟﻠﻐﺔ‬ ‫اﻟﻨﺼﻴﺔ‬ ‫ﻟﻠﻤﻌﺎﳉﺔ‬ ً ‫ﳎﺎﻧﺎ‬ ‫اﳌﺘﺎﺣﺔ‬ ‫ات‬ ‫ﻟﻸدو‬ ‫اﺳﻌﺔ‬ ‫و‬ ‫اﺳﺔ‬ ‫در‬ ‫ﺑﻌﻤﻞ‬ ‫أﻧﻪ‬ ‫ﺎ‬ ‫اﻵﱄ‬ ‫اﻟﺘﻌﻠﻢ‬ ‫ﺑﺮﳎﻴﺎت‬ ‫ﺗﻄﻮﻳﻊ‬ ‫ﳝﻜﻦ‬ ) machine learning ( ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫اﻟﻠﻐﺔ‬ ‫ﻋﻠﻰ‬ ‫ذﻟﻚ‬ ‫وﺗﻄﺒﻴﻖ‬ ‫ﻋﺎم‬ ‫ﺑﺸﻜﻞ‬ ‫اﻟﻠﻐﻮﻳﺔ‬ ‫ﻟﻠﻠﻤﻌﺎﳉﺔ‬ . ‫ﺑﻴﺔ‬ ‫ﻋﺮ‬ ‫ﻧﺼﻮص‬ ‫ﳎﻤﻮﻋﺔ‬ ‫أول‬ ‫ﲜﻤﻊ‬ ‫ﻗﻤﻨﺎ‬ ‫ﰒ‬ ‫وﻣﻦ‬ ‫اﻟﱪﳎﻴﺎت،‬ ‫ﻫﺬﻩ‬ ‫ﻳﺐ‬ ‫ﻟﺘﺪر‬ ‫ﻋﺮﰊ‬ ‫ﻧﺺ‬ ‫وﺟﻮد‬ ‫ﻳﺴﺘﺪﻋﻲ‬ ‫اﻷﻣﺮ‬ ‫وﻫﺬا‬ ) corpus ( ‫اﺠﻤﻟ‬ ‫ﻟﻠﺘﺤﻤﻴﻞ‬ ‫وﻣﺘﺎﺣﺔ‬ ‫اﳌﺼﺪر‬ ‫ﻣﻔﺘﻮﺣﺔ‬ ‫وﺟﻌﻠﻨﺎﻫﺎ‬ ‫ﻣﻨﺴﺠﻢ‬ ‫ﺑﺸﻜﻞ‬ ‫اﻟﻨﺺ‬ ‫ﻟﻌﺮض‬ ‫ﻣﺮﳛﺔ‬ ‫اﺟﻬﺔ‬ ‫و‬ ‫ﻧﺎﻣﺞ‬ ‫ﺑﺮ‬ ‫ﻧﺎ‬ ‫ﻃﻮر‬ ‫وﻛﺬﻟﻚ‬ ‫ﺎﱐ‬ . ‫اﺠﻤﻟﻤﻮﻋﺔ‬ ‫ﻫﺬﻩ‬ ‫اﻟﻨﺘﺎﺋﺞ‬ ‫وﺗﻘﻴﻴﻢ‬ ‫أﺧﺮى‬ ‫ﺣﺎﺳﻮﺑﻴﺔ‬ ‫اﻣﺞ‬ ‫ﺑﺮ‬ ‫ﻳﺐ‬ ‫ﻟﺘﺪر‬ ‫اﺳﺘﺨﺪﻣﻮﻫﺎ‬ ‫و‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫ﻟﻠﻐﺔ‬ ‫اﳊﺎﺳﻮﺑﻴﺔ‬ ‫اﳌﻌﺎﳉﺔ‬ ‫اء‬ ‫ﺧﱪ‬ ‫ﻗﺒﻞ‬ ‫ﻣﻦ‬ ‫اﺳﻌﺔ‬ ‫و‬ ‫ﺷﻬﺮة‬ ‫اﻛﺘﺴﺒﺖ‬ . ‫اﻟﺼ‬ ‫اﻟﺘﺤﻠﻴﻞ‬ ‫ﻣﺜﻞ‬ ‫اﳌﻌﺎﺻﺮ‬ ‫اﻟﻌﺮﰊ‬ ‫اﻟﻨﺺ‬ ‫ﻟﺘﺤﻠﻴﻞ‬ ‫ات‬ ‫أدو‬ ‫ﺑﺘﻄﻮﻳﺮ‬ ‫ﻗﻤﻨﺎ‬ ‫أﻧﻨﺎ‬ ‫ﻛﻤﺎ‬ ‫اﺳﻌﺔ‬ ‫اﻟﻮ‬ ‫ﻟﻠﺘﻐﻄﻴﺔ‬ ‫ﻟﻐﻮﻳﺔ‬ ‫وذﺧﺎﺋﺮ‬ ‫اﻟﻌﻨﻮﻧﺔ‬ ‫و‬ ‫اﻟﺘﺠﺬﻳﺮ‬ ‫ﺮﰲ،‬ ‫اﳋﻄﺎﺑﻴﺔ‬ ‫ﺑﺎﻟﻌﻼﻗﺎت‬ ‫ﻣﻌﻨﻮﻧﺔ‬ ‫ﻧﺼﻴﺔ‬ ‫ﳎﻤﻮﻋﺔ‬ ‫ﺗﻄﻮﻳﺮ‬ ‫و‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫ﻟﻠﻐﺔ‬ ) discourse relations ( ‫اﻟﻌﻼﻗﺎت‬ ‫ار‬ ‫ﻏﺮ‬ ‫ﻋﻠﻰ‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫ﻟﻠﻐﺔ‬ ‫ﺑﻨﺴﻠﻔﺎﻧﻴﺎ‬ ‫ﺟﺎﻣﻌﺔ‬ ‫ﻣﻦ‬ ‫اﻟﺼﺎدرة‬ ‫ﻳﺔ‬ ‫اﻹﳒﻠﻴﺰ‬ ‫ﻟﻠﻐﺔ‬ ‫اﳋﻄﺎﺑﻴﺔ‬ . ‫ﻧﻌﺘﺰ‬ ‫اﻟﱵ‬ ‫اﺋﺪة‬ ‫اﻟﺮ‬ ‫اﻟﺒﺤﺜﻴﺔ‬ ‫اﺠﻤﻟﺎﻻت‬ ‫وﻣﻦ‬ ‫ﻟﺘﻘﻨﻴﺔ‬ ‫إﻣﺘﺪاد‬ ‫وﺗﻌﺘﱪ‬ ‫اﻟﻜﺮﱘ‬ ‫ﻟﻠﻘﺮآن‬ ‫اﻟﻠﻐﻮﻳﺔ‬ ‫ﻟﻠﻤﻌﺎﳉﺔ‬ ‫ﺣﺎﺳﻮﺑﻴﺔ‬ ‫ﻳﻊ‬ ‫ﻣﺸﺎر‬ ‫ﻫﻮ‬ ‫ﻟﻴﺪز‬ ‫ﺟﺎﻣﻌﺔ‬ ‫ﰲ‬ ‫ﺎ‬ ‫اﻟﺘﻘﻠﻴﺪﻳﺔ‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫ﻟﻠﻐﺔ‬ ‫اﻟﻨﺼﻲ‬ ‫اﻟﺘﻨﻘﻴﺐ‬ . ‫ﻣﻮﻗﻊ‬ ‫و‬ ‫آﻟﻴﺔ‬ ‫ﳏﺎورة‬ ‫ﻧﺎﻣﺞ‬ ‫ﺑﺮ‬ ‫ﺗﻄﻮﻳﺮ‬ ‫اﻷﲝﺎث‬ ‫ﻫﺬﻩ‬ ‫وﻣﻦ‬ " ‫ﻗﺮآﱐ‬ " ‫ﻋﻠﻰ‬ ‫اﻟﺒﺤﺚ‬ ‫ﻳﺘﻴﺢ‬ ‫اﻟﺬي‬ ‫و‬ ‫اﻟ‬ ‫اﳌﻌﺮﻓﺔ‬ ‫ﻟﺘﻤﺜﻴﻞ‬ ‫وإﻃﺎر‬ ، ً ‫ﻣﺴﺒﻘﺎ‬ ‫ﻣﻌﺪة‬ ‫ﻣﻔﺎﻫﻴﻢ‬ ‫ﻣﺴﺘﻮى‬ ‫اﻟﻨﺤﻮﻳﺔ‬ ‫اﻟﻌﻨﻮﻧﺔ‬ ‫و‬ ‫ﻘﺮآﻧﻴﺔ‬ . ‫ﻣﻮﻗﻊ‬ ‫ﺑﺘﺪﺷﲔ‬ ‫ﻗﻤﻨﺎ‬ ً ‫ا‬ ‫وﻣﺆﺧﺮ‬ " ‫ﻟﻠﻘﺮآن‬ ‫ﺑﻴﺔ‬ ‫ﻋﺮ‬ ‫ﻟﻐﻮﻳﺔ‬ ‫ذﺧﲑة‬ ‫اﻟﻜﺮﱘ‬ ) " Quranic Arabic Corpus ] ( http //: corpus . quran . com [ ‫ﻟﻠﺘﺤﻤﻴﻞ‬ ‫ﻗﺎﺑﻞ‬ ‫إﻟﻜﱰوﱐ‬ ‫ﻣﻮرد‬ ‫وﻳﻌﺘﱪ‬ ‫ﻗﺮآﻧﻴﺔ‬ ‫ﻛﻠﻤﺔ‬ ‫ﻟﻜﻞ‬ ‫اﻟﻜﻼم‬ ‫أﻗﺴﺎم‬ ‫و‬ ‫اﻟﺼﺮف‬ ‫ﻣﺴﺘﻮى‬ ‫ﻋﻠﻰ‬ ‫ﺗﻔﺼﻴﻠﻴﺔ‬ ‫ﻟﻐﻮﻳﺔ‬ ‫ﲟﻌﻠﻮﻣﺎت‬ ‫وﻣﻮﺳﻢ‬ ‫اﺠﻤﻟﺎﱐ‬ . ‫وﻣﻨ‬ ‫ﺑﻘﺒﻮل‬ ‫ﺣﻈﻲ‬ ‫اﳌﻮﻗﻊ‬ ‫اﻧﻄﻼق‬ ‫ﺬ‬ ‫وﻫﻮ‬ ‫ﺑﻴﺔ،‬ ‫ﺑﺎﻟﻌﺮ‬ ‫اﻟﻨﺎﻃﻘﲔ‬ ‫ﻏﲑ‬ ‫ﻣﻦ‬ ‫اﻟﻘﺮآن‬ ‫ﻟﻐﺔ‬ ‫ﺗﻌﻠﻢ‬ ‫ﰲ‬ ‫اﻏﺒﲔ‬ ‫اﻟﺮ‬ ‫وﻛﺬﻟﻚ‬ ‫اﻟﻠﺴﺎﻧﻴﺎت‬ ‫أﲝﺎث‬ ‫ﰲ‬ ‫اء‬ ‫ﺧﱪ‬ ‫ﻣﻦ‬ ً ‫اﺳﻌﺎ‬ ‫و‬ ً ‫ا‬ ‫ﲨﻬﻮر‬ ‫اﺳﺘﻘﻄﺐ‬ ‫و‬ ‫اﺳﻊ‬ ‫و‬ ‫ﻣﻦ‬ ‫أﻛﺜﺮ‬ ‫وﳚﻠﺐ‬ ‫ﻟﻠﺒﺤﺚ‬ ‫ﺟﻮﺟﻞ‬ ‫ﳏﺮك‬ ‫ﻋﻠﻰ‬ ‫اﻟﻘﺮآن‬ ‫اﻋﺪ‬ ‫ﻗﻮ‬ ‫اﺳﺔ‬ ‫ﻟﺪر‬ ‫اﻷول‬ ‫اﳌﻮﻗﻊ‬ 50 ً ‫ﻳﺎ‬ ‫ﺷﻬﺮ‬ ‫اﺋﺮ‬ ‫ز‬ ‫أﻟﻒ‬ . ‫ﻧﻘﱰح‬ ‫ﺟﻌﻠﻨﺎ‬ ‫وﻫﺬا‬ " ‫ﻓﻬﻢ‬ ‫اﻟﻘﺮآن‬ " ‫ﻟﻌﺎم‬ ‫اﻟﺼﻨﺎﻋﻲ‬ ‫اﻟﺬﻛﺎء‬ ‫و‬ ‫اﳊﺎﺳﻮب‬ ‫ﻟﻌﻠﻮم‬ ‫ﻛﺒﲑ‬ ‫ﻛﺘﺤﺪ‬ 2010 ‫ﺑﻌﺪﻩ‬ ‫وﻣﺎ‬ . ‫أﻧﻨﺎ‬ ‫ﺣﻴﺚ‬ ‫اﳌﺸﺮوع‬ ‫ﻫﺬا‬ ‫ﺑﺘﻮﺳﻴﻊ‬ ‫ﻧﻘﻮم‬ ً ‫وﺗﺪرﳚﻴﺎ‬ ‫اﻟﻘﺮآﻧﻴﺔ‬ ‫اﳉﻤﻠﺔ‬ ‫ﻟﺒﻨﻴﺔ‬ ‫ﺗﺒﻌﻴﺔ‬ ‫ﺑﻴﺎﻧﻴﺔ‬ ‫ات‬ ‫ﺷﺠﺮ‬ ‫إﻋﺪاد‬ ‫ﻃﻮر‬ ‫ﰲ‬ ‫اﻵن‬ )) Quranic Arabic Dependency Treebank ‫ﻟﻠﻘﺮآء‬ ‫وﳑﻜﻨﺔ‬ ‫اﻟﻜﺮﱘ‬ ‫اﻟﻘﺮآن‬ ‫اب‬ ‫إﻋﺮ‬ ‫ﻛﺘﺐ‬ ‫ﻣﻦ‬ ‫ﻣﺴﺘﻤﺪة‬ ‫اﻵﻟﻴﺔ‬ ‫ة‬ . ‫ﺑﲔ‬ ‫اﺑﻂ‬ ‫رو‬ ‫ﺷﺒﻜﺔ‬ ‫وﺑﻨﺎء‬ ‫ﻗﺮآﻧﻴﺔ‬ ‫ﻣﻔﺎﻫﻴﻢ‬ ‫ﺑﺎﺳﺘﺨﻼص‬ ‫ﻧﻘﻮم‬ ‫أﻧﻨﺎ‬ ‫ﻛﻤﺎ‬ ‫اﳌﻔﺎﻫﻴﻢ‬ ‫ﳍﺬﻩ‬ ‫اﻟﻘﺮآن‬ ‫ﺿﻤﺎﺋﺮ‬ ‫وﻋﻮدة‬ ‫اﳌﻔﺎﻫﻴﻢ‬ ‫ﻫﺬﻩ‬ . ‫اﻟﻘﺮآن‬ ‫ﳌﻔﺮدات‬ ‫ﻟﻐﻮﻳﺔ‬ ‫ذﺧﲑة‬ ‫ﺑﻨﺎء‬ ‫اﻟﻘﺮآﱐ‬ ‫اﳌﺸﺮوع‬ ‫ﻫﺬا‬ ‫ﰲ‬ ‫اﳌﺴﺘﻘﺒﻠﻴﺔ‬ ‫اﳋﻄﻂ‬ ‫وﻣﻦ‬ ‫ﻣﺸﺮوع‬ ‫ار‬ ‫ﻏﺮ‬ ‫ﻋﻠﻰ‬ WordNet ‫ﳌﻔﺮدا‬ ‫دﻻﻟﻴﺔ‬ ‫ات‬ ‫إﻃﺎر‬ ‫ذﺧﲑة‬ ‫ﺗﻄﻮﻳﺮ‬ ‫وﻛﺬﻟﻚ‬ ‫ﻣﺸﺮوع‬ ‫ار‬ ‫ﻏﺮ‬ ‫ﻋﻠﻰ‬ ‫اﻟﻘﺮآن‬ ‫ت‬ FrameNet ‫ﻋﻠﻰ‬ ‫اﻟﻘﺮآﱐ‬ ‫اﻟﺒﺤﺚ‬ ‫ﻳﺘﻴﺢ‬ ‫ﻣﺘﻜﺎﻣﻞ‬ ‫ﲝﺚ‬ ‫ﳏﺮك‬ ‫ﺗﻄﻮﻳﺮ‬ ‫ﰒ‬ ‫وﻣﻦ‬ ‫اﳋﻄﺎﺑﻴﺔ،‬ ‫ﺑﺎﻟﻌﻼﻗﺎت‬ ‫ﻣﻌﻨﻮﻧﺔ‬ ‫ﻟﻠﻘﺮآن‬ ‫ﻧﺼﻴﺔ‬ ‫ﳎﻤﻮﻋﺔ‬ ‫ﺗﻄﻮﻳﺮ‬ ‫وﻛﺬﻟﻚ‬ ‫اﻟﺬﻛﺮ‬ ‫آﻧﻔﺔ‬ ‫اﻟﻠﻐﻮﻳﺔ‬ ‫اﻟﺬﺧﺎﺋﺮ‬ ‫ﺧﻼل‬ ‫ﻣﻦ‬ ‫وذﻟﻚ‬ ‫اﳌﻔﺎﻫﻴﻢ‬ ‫و‬ ‫اﻟﺼﺮف‬ ‫و‬ ‫اﻟﻨﺤﻮ‬ ‫و‬ ‫اﻷﺳﻠﻮب‬ ‫و‬ ‫اﳌﻔﺮدات‬ ‫ﻣﺴﺘﻮي‬ . ‫ﻧﺎﺣﺠﺔ‬ ‫ﻳﻊ‬ ‫ﻣﺸﺎر‬ ‫ﺧﻼل‬ ‫وﻣﻦ‬ ‫اﻟﺘﻘﻠﻴﺪﻳﺔ‬ ‫ﻟﻠﻨﺼﻮص‬ ‫اﳊﺎﺳﻮﺑﻴﺔ‬ ‫اﳌﻌﺎﳉﺔ‬ ‫ﰲ‬ ‫اﳌﻜﺘﺴﺒﺔ‬ ‫اﳋﱪة‬ ‫ﺗﻄﺒﻴﻖ‬ ‫ﻧﻄﻤﺢ‬ ‫اﻟﻜﺮﱘ‬ ‫اﻟﻘﺮآن‬ ‫ﻋﻠﻰ‬ -‫اث‬ ‫اﻟﱰ‬ ‫وﻛﺘﺐ‬ ‫اﻟﻨﺒﻮي‬ ‫اﳊﺪﻳﺚ‬ ‫ﻣﺜﻞ‬ ‫اﻟﻌﺮﰊ‬ -‫اﳌﻌﺎﺻﺮة‬ ‫اﻟﻔﺼﺤﻰ‬ ‫ﺑﻴﺔ‬ ‫اﻟﻌﺮ‬ ‫وﻛﺬﻟﻚ‬ . ‫اﻟﺘﻨﻘﻴﺐ‬ ‫و‬ ‫اﻟﺒﺤﺚ‬ ‫اض‬ ‫أﻏﺮ‬ ‫ﺗﻠﱯ‬ ‫ﻣﻌﻨﻮﻧﺔ‬ ‫ﻧﺼﻴﺔ‬ ‫ﳎﻤﻮﻋﺔ‬ ‫ﻟﺘﻄﻮﻳﺮ‬ ‫ﳔﻄﻂ‬ ‫اﳌﺜﺎل‬ ‫ﺳﺒﻴﻞ‬ ‫ﻓﻌﻠﻰ‬ ‫اﻹ‬ ‫ﳎﺎل‬ ‫ﰲ‬ ‫ﻟﻠﻤﺨﺘﺼﲔ‬ ‫اﻟﻨﺼﻲ‬ ‫اﻹﺧﺘﺼﺎص‬ ‫ﻫﺬا‬ ‫ﰲ‬ ‫اﳌﻌﻨﻮﻧﺔ‬ ‫ﻟﻠﻤﻔﺎﻫﻴﻢ‬ ‫اﳌﻌﺮﰲ‬ ‫اﻟﺘﻤﺜﻴﻞ‬ ‫ﺧﻼل‬ ‫ﻣﻦ‬ ‫وذﻟﻚ‬ ‫اﻹﺳﻼﻣﻲ‬ ‫ﻗﺘﺼﺎد‬ . ‫وﻛﺬﻟﻚ‬ ‫ﺗﻘﻨﻴﺎت‬ ‫ﺗﻮﻇﻴﻒ‬ ‫ﻣﻦ‬ ‫ﳝﻜﻨﻨﺎ‬ ‫وﻫﺬا‬ ‫اﻹﳒﻴﻞ‬ ‫و‬ ‫اة‬ ‫اﻟﺘﻮر‬ ‫ﻣﺜﻞ‬ ‫اﻷﺧﺮى‬ ‫اﻷدﻳﺎن‬ ‫ﰲ‬ ‫ﻣﻘﺪﺳﺔ‬ ‫ﻛﺘﺐ‬ ‫ﻟﺘﺸﻤﻞ‬ ‫اﳊﺎﺳﻮﺑﻴﺔ‬ ‫أﲝﺎﺛﻨﺎ‬ ‫ﲤﺪﻳﺪ‬ ‫ﻋﻠﻴﻨﺎ‬ ‫ﺮض‬ ُ ‫ﻋ‬ ‫اﳌﺼﺎدر‬ ‫ﻫﺬﻩ‬ ‫ﺑﲔ‬ ‫اﻟﻔﺮوﻗﺎت‬ ‫و‬ ‫اﻟﺘﺸﺎﺑﻪ‬ ‫ﳌﻼﺣﻈﺔ‬ ‫ﺣﺎﺳﻮﺑﻴﺔ‬ .

Statistical Machine Translation (SMT) is considered as sub-field of computational linguistics; and
the latter is regarded as a branch of Artificial Intelligence (AI) dedicated to Natural Language Processing
(NLP). The main purpose of this paper is shortening the distance between the Language and the most recent
cutting edge technology dedicated to Machine Translation (MT). On the one hand, Statistical Machine
Translation (SMT) considers the translation as a human craft, hence it uses linguistics monolingual and
bilingual corpora translated by professional translators; the monolingual are used to train the Language
Models (LM) and the bilingual are used to train the Translation Models (TM). On the other hand, it takes
advantage of the processing high performance of computers by integrating Statistical Methods to select the
best translation. This paper presents the basic concepts and approaches of Machine Translation (MT), and
focuses on SMT, then introducing the features of the open source Moses Decoder. This system has been
experimented and adapted to translate from English into Arabic. The English-Arabic prototype has been
evaluated using MEDAR MT package and the obtained results were very encouraging.

Plagiarism detection means detecting the document whether copied or stealing from the other document.
The main goal is to detect the word by analyzing the writing style using technique intrinsic plagiarism detection.
Text mining is used to extract the useful information from the text. Intrinsic plagiarism detection is used to take the
few words from the document and then it compared to the original document whether it’s plagiarized or not. In
addition, it performs the modern method in intrinsic plagiarism detection such as Recall, Precision, F-measure and
Granularity. stylometric frequent pattern of authors can be extracted and mined along with lexical character features,
lexical word features and syntactical features. The proposed work detects the plagiarism for documents based on
collecting author profiles of documents.

This paper describes a new tool that helps extracting clean text from the Arabic Wikisource dump in order to build corpora. The tool purpose is illustrated by the generation of a subcorpus from Wikisource, which is a step towards the building of an evaluation corpus for Arabic intrinsic plagiarism detection.