Arabic open information extraction system using dependency parsing

<p>Arabic is a Semitic language and one of the most natural languages distinguished by the richness in morphological enunciation and derivation. This special and complex nature makes extracting information from the Arabic language difficult and always needs improvement. Open information extraction systems (OIE) have been emerged and used in different languages, especially in English. However, it has almost not been used for the Arabic language. Accordingly, this paper aims to introduce an OIE system that extracts the relation tuple from Arabic web text, exploiting Arabic dependency parsing and thinking carefully about all possible text relations. Based on clause types' propositions as extractable relations and constituents' grammatical functions, the identities of corresponding clause types are established. The proposed system named Arabic open information extraction(AOIE) can extract highly scalable Arabic text relations while being domain independent. Implementing the proposed system handles the problem using supervised strategies while the system relies on unsupervised extraction strategies. Also, the system has been implemented in several domains to avoid information extraction in a specific field. The results prove that the system achieves high efficiency in extracting clauses from large amounts of text.</p>

The Causative Stem in Soqotri

Soqotri is an understudied Semitic language belonging to the Modern South Arabian branch and spoken by the approximately 100,000 inhabitants of the island of Soqotra. The present contribution offers a list of verbs belonging to the so-called causative stem in Soqotri (a cognate of the Arabic stem IV), based on the analysis of the data in the two recently published volumes of the Soqotri oral literature as well as the fieldwork notes of the authors.

Introducing Various Semantic Models for Amharic: Experimentation and Evaluation with Multiple Tasks and Datasets

The availability of different pre-trained semantic models has enabled the quick development of machine learning components for downstream applications. However, even if texts are abundant for low-resource languages, there are very few semantic models publicly available. Most of the publicly available pre-trained models are usually built as a multilingual version of semantic models that will not fit well with the need for low-resource languages. We introduce different semantic models for Amharic, a morphologically complex Ethio-Semitic language. After we investigate the publicly available pre-trained semantic models, we fine-tune two pre-trained models and train seven new different models. The models include Word2Vec embeddings, distributional thesaurus (DT), BERT-like contextual embeddings, and DT embeddings obtained via network embedding algorithms. Moreover, we employ these models for different NLP tasks and study their impact. We find that newly-trained models perform better than pre-trained multilingual models. Furthermore, models based on contextual embeddings from FLAIR and RoBERTa perform better than word2Vec models for the NER and POS tagging tasks. DT-based network embeddings are suitable for the sentiment classification task. We publicly release all the semantic models, machine learning components, and several benchmark datasets such as NER, POS tagging, sentiment classification, as well as Amharic versions of WordSim353 and SimLex999.

Defining Definiteness in Turoyo

This paper puts forth an expanded typology of definiteness marking, which includes not only ‘strong’ and ‘weak’ but also ‘super-weak’. It also proposes a methodology for identifying ‘super-weak’ definites, and applies it to Ṭuroyo, an endangered Semitic language. Data from questionnaires and interviews shows that Ṭuroyo’s definite article has a very wide distribution, including anti-uniqueness effects with exclusives, suggesting ‘super-weak’ status. Syntactic factors also affect their distribution: We find definiteness-spreading uses with demonstratives and possessives, even in non-contrastive environments, and superlative adjectives appear to compete for the article’s syntactic position. On the semantic side, we propose that Ṭuroyo’s definiteness-markers are not ‘weak’ but ‘super-weak’ articles. To explain their anaphoric uses, typical of ‘strong’ articles, we propose that the typology is arranged as a cline ordered by entailment, so that stronger articles entail weaker ones.

Kronologi Bahasa Arab Semitik Perspektif Historis

Arabic comes from the Semitic language family. it has a higher position than other Semitic languages ​​because it is not extinct even though it is thousands of years old. Arabic is one of the most widely spoken languages ​​today. Users are not only Muslims but also non-Muslims. Semitic is a language family associated with one of the sons of the Prophet Noah a.s. namely Sam son of Noah. In the narration, it is stated that Noah had three children, namely Sam (Semitic), Yafit (Aramiyah), and Ham (Hamiyah). Languages ​​belonging to the Semitic language family experience slow development. Some languages, such as Akkadian, have even become extinct. and other languages, although not extinct but the speakers are no longer a big nation. This language is only spoken by a small number of people, like Aramaic. Only Arabic, the only Semitic language, has survived thousands of years. Arabic users continue to grow even today. The final conclusion of this paper is that Arabic represents other Semitic languages, both in terms of their origin and in terms of their privilege.

Building a Dataset for Detecting Fake News in Amharic Language

The rapid growth and expansion of social media platform has filled the gap of information exchange in the day to day life. Apparently, social media is the main arena for disseminating manipulated information in a high range and exponential rate. The fabrication of twisted information is not limited to ones language, society and domain, this is particularly observed in the ongoing COVID-19 pandemic situation. The creation and propagation of fabricated news creates an urgent demand for automatically classification and detecting such distorted news articles. Manually detecting fake news is a laborious and tiresome task and the dearth of annotated fake news dataset to automate fake news detection system is still a tremendous challenge for low-resourced Amharic language (after Arabic, the second largely spoken Semitic language group). In this study, Amharic fake news dataset are crafted from verified news sources and various social media pages and six different machine learning classifiers Naïve bays, SVM, Logistic Regression, SGD, Random Forest and Passive aggressive Classifier model are built. The experimental results show that Naïve bays and Passive Aggressive Classifier surpass the remaining models with accuracy above 96% and F1- score of 99%. The study has a significant contribution to turn down the rate of disinformation in vernacular language.

The Semantics of Locative Adpositions in Amharic

This paper intends to describe the semantics of locative adpositions in Amharic, a Semitic language spoken in Ethiopia. The analysis is based on elicited data that were collected by using Bowerman and Pederson’s (1992) topological relations picture series. The study shows that Amharic locative adpositions can convey specific and generic topological relations between the figure and ground entities. The specific locatives show a specific type of topological relations (for instance, verticality as in tatʃtʃ, below’, ‘under’; horizontality as in fit, ‘front’; containment as in wɨst’, ‘in’) between the figure and ground entities, but not the generic locatives. Aside from which, I argue that Amharic does not fit into Ameka and Levinson’s (2007) typology of locative predicates and constitutes a type of its own because it uses two copulas and a locative verb.

Evaluating the Translation of Accusative of Purpose by EFL learners

Arabic is the most widely spoken Semitic language and has seen phases of change from the Quranic form to the more popular Modern Standard Arabic that is used for communication today. Teachers of language and translation in KSA are concerned with learning problems that arise due to lexical and grammatical non-equivalence between Arabic and English which often leads to confusion and incorrect output during translation process. The current study aimed at investigating one of the Arabic grammatical structures which has no direct equivalent in English (Accusative of Purpose). It is an original noun derived from a verb of affectivity, explaining the reason why the action of the previous verb has been done. Two objectives were targeted: One, testing the learners’ ability to translate the Accusative of Purpose; and two, to gather an understanding of the strategies they adopted in the process. The study is likely to be of great value in a foreign language learning environment as is the case in the KSA. We used written tests to collect the data, followed by detailed interviews to elicit information on the translation strategies used. Participants were female undergraduate students (N=35) at Hurimilla College of Science and Humanities, Shaqra University, KSA, of which fifteen were randomly interviewed consequently. Both qualitative and quantative approaches were used. The data collected was analysed using SPSSR. The results revealed that (36.0%) of the students’ translations were correct or acceptable, while 34.3% were literal, and (15.4%) were incorrect. Personal interviews revealed that reasons of these results can be attributed to the absence of this category in English, and non-equivalence between Arabic and English. Based on these findings, the present study suggests more investigations on similar Arabic structures which have no correspondence in English using larger number of participants.

Jeremiah and Baruch

This chapter discusses two books that are inseparable in the Septuagint manuscripts: Jeremiah and Baruch. For both books, it surveys classical topics like translators, translation technique, and provenance. The main focus, however, is on the manner they exemplify important and recurrent issues in Septuagintal studies. The book of Jeremiah is a textbook case of a work that exists in two literary editions, a short one (in Hebrew) and a long one (in Greek). The study of the book of Baruch is a good example of a pseudepigraphical work that seems to be translated from a Semitic language, although this is disputed.

Die Modalverben im Deutschen und Arabischen

ان مقارنة قواعد اللغة الألمانية مع لغة سامية كاللغة العربية تفضي الى حجم الهوة والاختلافات الكبيرة بين قواعد وأنظمة اللغة العربية مقارنة باللغة الألمانية. فموضوع البحث يتناول مقارنة افعال الكيفية في اللغة الألمانية وما يقابها باللغة العربية. أفعال الكيفية هي عبارة عن ستة الفعال لها قواعد تركيبية خاصة تختلف عن الأفعال الأخرى من حيث البناء والتركيب، وهي أفعال شاذة ولها تصاريف معينة مع الضمائر الشخصية سواء في الزمن الماضي او المضارع، فهي تستلزم غالبا ما فعل مصدري يكتب في نهاية الجملة لاتمام المعنى، ويعد هذا الفعل فعلا رئيسيا، وتأخذ هذه الأفعال الموضع الثاني في الجملة وتكون مصرفة حسب الفاعل بينما يأتي الفعل الرئيسي في نهاية الجملة دون أي تصريف، وهذه الأفعال هي: يَسمَح، يَستطيع، يَوَد أو يَرغَب، يَجب، يَنبغي، يُريد. الفعل في قواعد اللغة العربية يدل على معنى في نفسه فمنه المعرب ومنه المبني وله اوزان محددة، والفعل الثلاثي الماضي على وزن فَعَلَ تعد أساسا للميزان الصرفي، فعند ترجمة أفعال الكيفية حسب الميزان الصرفي تكون أرادَ وانبغى واستطاع .. الخ. أفعال الكيفية في اللغة الألمانية تقابلها في اللغة العربية أنْ المصدرية الناصبة والتي تدخل على الجمل الفعلية وتفيد الحال والاستقبال اذا دخلت على الفعل المضارع. وقد تترجم الجمل والنصوص الى اللغة العربية بالصدر المؤول من (أنْ والفعل) او بالمصدر الصريح، وبعضها يرتبط بحرف جر مثل سُمِحَ( لَه)ُ و وَجَبَ (على)، وهناك الكثير من المفردات والاستخدامات اللغوية تم ذكرها بالبحث. Comparing the grammar of the German language with a Semitic language such as the Arabic language leads to the size of the gap and the big differences between the rules and systems of the Arabic language compared to the German language. The topic of this research deals with comparing the qualitative verbs in the German language and the equivalent in the Arabic language. Modal verbs are six active verbs that have special syntactic rules that differ from other verbs in terms of construction and composition. They are irregular verbs and have specific conjugations with personal pronouns, whether in the past or present tense, they often require an infinitive verb written at the end of the sentence to complete the meaning. This verb is a main verb, and these verbs take the second position in the sentence and are inflected according to the subject, while the main verb comes at the end of the sentence without any conjugation, and these verbs are: allow, can, wish or desire, should, should, want. The verb in the grammar of the Arabic language denotes a meaning in itself, from which it is expressed, and from it that is based and has specific weights, and the past triple verb on the weight of فَعَلَ is the basis for the morphological balance, so when translating modal verbs according to the morphological scale, they are أراد ، انبغى and أستطاعَ ... etc. Verbs of modal in the German language correspond to it in the Arabic language that the أنْ accusative infinitive, which is entered into the verbal sentences, and benefits the adverb and the reception if entered into the present tense. Sentences and texts may be translated into Arabic with the source of (أنْ and verb) or with the explicit source, and some of them are related to a preposition such as allowed (for him) and obligatory (on), and there are many vocabulary and linguistic uses mentioned in the research.