Towards an Arabic Sign Language (ArSL) corpus for deaf drivers

Sign language is a common language that deaf people around the world use to communicate with others. However, normal people are generally not familiar with sign language (SL) and they do not need to learn their language to communicate with them in everyday life. Several technologies offer possibilities for overcoming these barriers to assisting deaf people and facilitating their active lives, including natural language processing (NLP), text understanding, machine translation, and sign language simulation. In this paper, we mainly focus on the problem faced by the deaf community in Saudi Arabia as an important member of the society that needs assistance in communicating with others, especially in the field of work as a driver. Therefore, this community needs a system that facilitates the mechanism of communication with the users using NLP that allows translating Arabic Sign Language (ArSL) into voice and vice versa. Thus, this paper aims to purplish our created dataset dictionary and ArSL corpus videos that were done in our previous work. Furthermore, we illustrate our corpus, data determination (deaf driver terminologies), dataset creation and processing in order to implement the proposed future system. Therefore, the evaluation of the dataset will be presented and simulated using two methods. First, using the evaluation of four expert signers, where the result was 10.23% WER. The second method, using Cohen’s Kappa in order to evaluate the corpus of ArSL videos that was made by three signers from different regions of Saudi Arabia. We found that the agreement between signer 2 and signer 3 is 61%, which is a good agreement. In our future direction, we will use the ArSL video corpus of signer 2 and signer 3 to implement ML techniques for our deaf driver system.

NEUROPSYCHOLOGICAL DETERMINANTS OF DYSLEXIA IN SCHOOLCHILDREN

The article describes the results of the research exploring the link between the neurodynamic component of schoolchildren’s psychic activity at the age of 7-12 and the level of mastering the reading skill at school. In the process of the research, we have detected the core neuropsychological criteria correlating with the level of reading technique and text understanding in schoolchildren. The neuropsychological research was made using A.R. Luria’s battery of tests, by G.M. Glosman’s method, adapted to childhood. We also used A.N. Kornyev and O.A. Ishymova’s method of dyslexia diagnostics; texts for the research have been translated into Ukrainian. We have pointed out two age groups of children andanalyzed how neurodynamic deficiency and audio-verbal and visual memory level influence the formation of dyslexia determinants in both of them. In addition, we have addressed the link between child’s development under the age of 1, parents’ complaints at the actual time and the level of a child’s reading technique.

Weakly Supervised Learning for Textbook Question Answering

Textbook Question Answering (TQA) is the task of answering diagram and non-diagram questions given large multi-modal contexts consisting of abundant text and diagrams. Deep text understandings and effective learning of diagram semantics are important for this task due to its specificity. In this paper, we propose a Weakly Supervised learning method for TQA (WSTQ), which regards the incompletely accurate results of essential intermediate procedures for this task as supervision to develop Text Matching (TM) and Relation Detection (RD) tasks and then employs the tasks to motivate itself to learn strong text comprehension and excellent diagram semantics respectively. Specifically, we apply the result of text retrieval to build positive as well as negative text pairs. In order to learn deep text understandings, we first pre-train the text understanding module of WSTQ on TM and then fine-tune it on TQA. We build positive as well as negative relation pairs by checking whether there is any overlap between the items/regions detected from diagrams using object detection. The RD task forces our method to learn the relationships between regions, which are crucial to express the diagram semantics. We train WSTQ on RD and TQA simultaneously, \emph{i.e.}, multitask learning, to obtain effective diagram semantics and then improve the TQA performance. Extensive experiments are carried out on CK12-QA and AI2D to verify the effectiveness of WSTQ. Experimental results show that our method achieves significant accuracy improvements of $5.02\%$ and $4.12\%$ on test splits of the above datasets respectively than the current state-of-the-art baseline. We have released our code on \url{https://github.com/dr-majie/WSTQ}.

Weakly Supervised Learning for Textbook Question Answering

Textbook Question Answering (TQA) is the task of answering diagram and non-diagram questions given large multi-modal contexts consisting of abundant text and diagrams. Deep text understandings and effective learning of diagram semantics are important for this task due to its specificity. In this paper, we propose a Weakly Supervised learning method for TQA (WSTQ), which regards the incompletely accurate results of essential intermediate procedures for this task as supervision to develop Text Matching (TM) and Relation Detection (RD) tasks and then employs the tasks to motivate itself to learn strong text comprehension and excellent diagram semantics respectively. Specifically, we apply the result of text retrieval to build positive as well as negative text pairs. In order to learn deep text understandings, we first pre-train the text understanding module of WSTQ on TM and then fine-tune it on TQA. We build positive as well as negative relation pairs by checking whether there is any overlap between the items/regions detected from diagrams using object detection. The RD task forces our method to learn the relationships between regions, which are crucial to express the diagram semantics. We train WSTQ on RD and TQA simultaneously, \emph{i.e.}, multitask learning, to obtain effective diagram semantics and then improve the TQA performance. Extensive experiments are carried out on CK12-QA and AI2D to verify the effectiveness of WSTQ. Experimental results show that our method achieves significant accuracy improvements of $5.02\%$ and $4.12\%$ on test splits of the above datasets respectively than the current state-of-the-art baseline. We have released our code on \url{https://github.com/dr-majie/WSTQ}.

A General Method for Transferring Explicit Knowledge into Language Model Pretraining

Recently, pretrained language models, such as Bert and XLNet, have rapidly advanced the state of the art on many NLP tasks. They can model implicit semantic information between words in the text. However, it is solely at the token level without considering the background knowledge. Intuitively, background knowledge influences the efficacy of text understanding. Inspired by this, we focus on improving model pretraining by leveraging external knowledge. Different from recent research that optimizes pretraining models by knowledge masking strategies, we propose a simple but general method to transfer explicit knowledge with pretraining. To be specific, we first match knowledge facts from a knowledge base (KB) and then add a knowledge injunction layer to a transformer directly without changing its architecture. This study seeks to find the direct impact of explicit knowledge on model pretraining. We conduct experiments on 7 datasets using 5 knowledge bases in different downstream tasks. Our investigation reveals promising results in all the tasks. The experiment also verifies that domain-specific knowledge is superior to open-domain knowledge in domain-specific task, and different knowledge bases have different performances in different tasks.

Towards a Theology of Migration: A Survival Perspective from Isaiah 1-12

The aim of this paper is to turn the lens towards the pre-migration and resilience processes within the context of the imminent forced migration as found in Isaiah 1-12. The article addresses not only the matter of pre-migration and collective trauma but also the ensuing resilience and hope that is embedded in the text. Understanding the concepts that underpin premigration to trauma and hope, the authors have engaged Isaiah 1-12, which present substantial pre-migration trauma markers of collective trauma, resilience and hope in the text. This article offers original research in the field of pre-migration trauma studies in Isaiah 1-12, as very little research has been done on this topic. An attempt was made to start a new conversation and understanding about forced migration and trauma within the field of Isaiah 1-12 and biblical studies.

Poetic discourse: text, hermeneutics, harmony

The article is devoted to the investigation of poetic discourse. Discourse is understood as the actualization of text structures in interaction with extralinguistic factors. The article considers the problems of understanding, methods of comprehension and the method of analyzing a poetic text; understanding of the unity of language and thinking, the author's axiological paradigm, refracted through the prism of the dynamics of cultural experience. The author analyze fiction that allow to retrace the formation of poetic concept. The influence of the structure on the formation of text is also presented. The main individual concepts are determined. The given examples illustrate the statements under consideration.

IMPACT OF MACHINE TRANSLATION ON UNIVERSITY STUDENTS’ TRANSLATION ABILITY: SOUTH EASTERN UNIVERSITY OF SRI LANKA AS A MODEL

Machine Translation (MT) is a unique tool in the field of translation. It is used all over the world. In accordance with, the undergraduates who are following Specialization in Linguistics and Translation at the department of Arabic language from South Eastern University of Sri Lanka face obstacles when they translate between Tamil, English & Arabic. Due to, they focus on MT to translate. According to this, the current research aims to identify the impact of MT on their translation activities. This research uses the descriptive analysis methodology. Primary data were collected from observation and questionnaire that was given to the undergraduates who are following Specialization in Linguistics and Translation at the department of Arabic language from South Eastern University of Sri Lanka in the academic year 2018/2019. Secondary data were gathered from research papers, books, research articles, and websites. The findings of this research declares that the undergraduates in the selected area have desires in the translation field and face the obstacles when translating. Thus, they try to use MT for getting accurate translation, idea of the source text, understanding the unknown words. Due to it, they couldn’t omit the MT in their translation activities. Because, it is easy to use and helps to save the time. At the time, they assume that non-use of MT will effect negatively in their translation ability.