Showing papers in "Journal of Computer Science in 2020"

A Survey of Methods for Managing the Classification and Solution of Data Imbalance Problem

Abstract: The problem of class imbalance is extensive for focusing on numerous applications in the real world. In such a situation, nearly all of the examples are labeled as one class called majority class, while far fewer examples are labeled as the other class usually, the more important class is called minority. Over the last few years, several types of research have been carried out on the issue of class imbalance, including data sampling, cost-sensitive analysis, Genetic Programming based models, bagging, boosting, etc. Nevertheless, in this survey paper, we enlisted the 24 related studies in the years 2003, 2008, 2010, 2012 and 2014 to 2019, focusing on the architecture of single, hybrid and ensemble method design to understand the current status of improving classification output in machine learning techniques to fix problems with class imbalances. This survey paper also includes a statistical analysis of the classification algorithms under various methods and several other experimental conditions, as well as datasets used in different research papers.

GoogleNet CNN neural network towards chest CT-coronavirus medical image classification

Abstract: In the end of the year 2019 and the beginning of the year 2020, the world was overwhelmed by a medical pandemic that was not previously seen which is known Covid-19 (Coronavirus) Coronavirus (CoV) is a large family of viruses that cause illness ranging from the common cold to more severe diseases such as Middle East Respiratory Syndrome (MERS-CoV) and Severe Acute Respiratory Syndrome (SARS-CoV) This paper aims to improve the accuracy of detection for CT-Coronavirus images using deep learning for Convolutional Neural Networks (CNNs) that helps medical staffs for classification chest CT-Coronavirus medical image in early stage Deep learning is successfully used as a tool for machine learning, where the CNNs are capable of automatically extracting and learning features medical image dataset This research retrains GoogleNet CNN architecture over the COVIDCT-Dataset for classification CT-Coronavirus image In this research, COVIDCT-Dataset contains 349 CT images containing clinical findings of COVID-19 The validation accuracy of retraining GoogleNet is 82 14% where elapsed time is 74 min and 37 sec © 2020 Nesreen Alsharman and Ibrahim Jawarneh

General Approach for Formulating a Digital Transformation Strategy

Abstract: To take advantage of information technologies, organizations need to define a clear strategy. Numerous works have provided definitions and phases of digital strategies. Some of these strategies are context-specific, while others deal with the common elements of digital strategies regardless of the digital transformation context. However, these works do not address a holistic approach. This raises ambiguity regarding digital strategy definitions and approaches. To eliminate this ambiguity, the current research tries to take advantage of existing digital strategies to propose a general digital strategy definition and build a general digital transformation approach. This work analyses various digital transformation strategies, to extract and classify their common elements in order to build a general approach that frame and drive the formulation of digital transformation strategies. To define such a general approach, the current paper analyses the effects of IT Governance and Management Strategy on the Digital Transformation Maturity. This analysis identified how IT Governance and Management Strategy can contribute to formulating a digital transformation strategy. Partial Least Square (PLS) was adopted in this research to develop an empirical evaluation for the case of 30 digital strategies and frameworks. Based on this empirical study several results have been presented in this work, namely: determination of a digital strategy definition and identification of a digital strategy approach. The proposed approach is composed of the following building blocks: Strategic Awareness, Business Strategic Planning, IT Organizational Structure, Steering committee, IT Prioritization Process, IT Investment Decisions, IT Strategic Planning, IT Budgeting, IT Reporting, IT Reaction Capacity and Management Strategy.

A Survey on Event Detection Models for Text Data Streams

Abstract: Event Detection (ED) is a study area that attracts the attention of decision-makers from various disciplines in order to help them in taking the right decision. ED has been examined on various text streams like Twitter, Facebook, Emails, Blogs, Web Forums and newswires. Many ED models have been proposed in literature. In general, ED model consists of six main phases: Data collection, pre-processing, feature selection, event detection, performance evaluation and result representation. Among these phases, event detection phase has a vital rule in the performance of the ED model. Consequently, numerous supervised, unsupervised, semi-supervised detection methods have been introduced for this phase. However, unsupervised methods have been extensively utilized as ED process is considered as unsupervised task. Hence, such methods need to be categorized on such a way so it can help researchers to understand and identified the limitations lay in these methods. In this survey, ED models for text data from various Social Network sites (SNs) are analyzed based on domain type, detection methods, type of detection task. In addition, main categories for unsupervised detection methods are explicitly mentioned with revising their related works. Moreover, the major open challenges faced by researchers for building ED models are explained and discussed in detail. The main objective of this survey paper is to provide a complete view of the recent developments in ED field. Hence, help scholars to identify the limitations of existing ED models for text data and help them to recognize the interesting future works directions.

Water Resource Evaluation and Identifying Groundwater Potential Zones in Arid area using Remote Sensing and Geographic Information System

Abstract: Groundwater resource is the main conventional source of fresh water all over the world. However, recent revelations indicated that the shortage of water resources remains the main challenge for the arid areas. In this regard, identifying groundwater potential zones or areas can help to improve the availability of fresh water and effective management of groundwater in arid areas. This work finds the water resources and identify the groundwater potential zones of arid areas using remote sensing and GIS techniques. The study uses Kingdom of Saudi Arabia (KSA) as one of the most arid area and divides entire KSA into five regions namely northern, central, western, southern and eastern to evaluate and indicate the groundwater prospective zones effectively and clearly. The northern region (Al Jouf, Tabuk, Hail and Al-Qassim), Saq and overlying aquifers play an important role in water supply in Saudi Arabia. About 17.90% of the total area of this region identified as a groundwater potential zone. Based on geomorphological factors, the Wadi catchment areas act as the best appropriate regions for groundwater recharge in the northern area. Regarding the central region (Al-Riyad province), about 1.47% and 4.15% may be categorized as excellent and very good while 12.59%, 74.82% and 6.97% are considered as good, poor and very poor groundwater potential zone. In the western area (Wadi Yalanlan basin), the lower part of the Wadi Yalamlam basin is the most promising zone for groundwater availability containing both high and moderate potential areas. Also, high groundwater potential zones can be found on the northern side of the central dyke region surrounding Abu Helal’s farm. 50.5% and 31% of the southern area (Jazan region) contain excellent and good groundwater potential areas while 16% and 2.5% of the regions showed average low groundwater potential zones. The eastern region had characteristics of extreme arid and desert environments. Based on the features, the area did not contain any groundwater potential zone. The current evaluation of groundwater potential areas in Saudi Arabia can serve as a significant tool for efficient groundwater resource management.

Multipatch-GLCM for Texture Feature Extraction on Classification of the Colon Histopathology Images using Deep Neural Network with GPU Acceleration

Abstract: Cancer is one of the leading causes of death in the world. It is the main reason why research in this field becomes challenging. Not only for the pathologist but also from the view of a computer scientist. Hematoxylin and Eosin (H&E) images are the most common modalities used by the pathologist for cancer detection. The status of cancer with histopathology images can be classified based on the shape, morphology, intensity, and texture of the image. The use of full high-resolution histopathology images will take a longer time for the extraction of all information due to the huge amount of data. This study proposed advance texture extraction by multi-patch images pixel method with sliding windows that minimize loss of information in each pixel patch. We use texture feature Gray Level Co-Occurrence Matrix (GLCM) with a mean-shift filter as the data pre-processing of the images. The mean-shift filter is a low-pass filter technique that considers the surrounding pixels of the images. The proposed GLCM method is then trained using Deep Neural Networks (DNN) and compared to other classification techniques for benchmarking. For training, we use two hardware: NVIDIA GPU GTX-980 and TESLA K40c. According to the study, Deep Neural Network outperforms other classifiers with the highest accuracy and deviation standard 96.72±0.48 for four cross-validations. The additional information is that training using Theano framework is faster than Tensorflow for both in GTX-980 and Tesla K40c.

Thinging-Based Conceptual Modeling: Case Study of a Tendering System

Abstract: In computer science, models are made explicit to provide formality and a precise understanding of small, contingent universes (e.g., an organization), as constructed from stakeholder requirements. Conceptual modeling is a fundamental discipline in this context whose main concerns are identifying, analyzing and describing the critical concepts of a universe of discourse. In the information systems field, one of the reasons why projects fail is an inability to capture requirements in a way that can be technically used to configure a system. This problem of requirements specification is considered to have deficiencies in theory. We apply a recently developed model called the Thinging Machine (TM) model which uniformly integrates static and dynamic modeling features to this problem of requirements specification. The object-Oriented (OO) approach to modeling, as applied in Unified Modeling Language, is by far the most applied and accepted standard in software engineering; nevertheless, new notions in the field may enhance and facilitate a supplementary understanding of the OO model itself. We aim to contribute to the field of conceptual modeling by introducing the TM model s philosophical foundation of requirements analysis. The TM model has only five generic processes of things (e.g., objects), in which genericity indicates generality, as in the generic Aristotelian concepts based on abstraction. We show the TM model s viability by applying it to a real business system.

How to Document Computer Networks

Abstract: Documenting networks is an essential tool for troubleshooting network problems The documentation details a network's structure and context, serves as a reference and makes network management more effective Complex network diagrams are hard to document and maintain and are not guaranteed to reflect reality They contain many superficial icons (eg, wall, screen and tower) Defining a single coherent network architecture and topology, similar to engineering schematics, has received great interest We propose a fundamental approach for methodically specifying a network architecture using a diagramming method to conceptualize the network s structure The method is called a thinging (abstract) machine, through which the network world is viewed as a single unifying element called the thing/machine (thimac), providing the ontology for modeling the network To test its viability, the thinging-machine-based methodology was applied to an existing computer network to produce a single integrated, diagrammatic representation that incorporates communication, software and hardware The resultant description shows a viable, coherent depiction that can replace the current methods

Prediction of Data Traffic in Telecom Networks based on Deep Neural Networks

Abstract: Corresponding Author: Quang Hung Do Faculty of Information Technology, University of Transport Technology, Vietnam Email: hungdq@utt.edu.vn Abstract: Accurate prediction of data traffic in telecom network is a challenging task for a better network management. It advances dynamic resource allocation and power management. This study employs deep neural networks including Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU) techniques to one-hour-ahead forecast the volume of expected traffic and compares this approach to other methods including Adaptive Neuro-Fuzzy Inference System (ANFIS), Artificial Neural Network (ANN) and Group Method of Data Handling (GMDH). The deep neural network implementation in this study analyses, evaluates and generates predictions based on the data of telecommunications activity every one hour, continuously in one year, released by Viettel Telecom in Vietnam. The performance indexes, including RMSE, MAPE, MAE, R and Theil’s U are used to make comparison of the developed models. The obtained results show that both LSTM and GRU model outperformed the ANFIS, ANN and GMDH models. The research findings are expected to provide an assistance and forecasting tool for telecom network operators. The experimental results also indicate that the proposed model is efficient and suitable for real-world network traffic prediction.

Modeling the Semantics of States and State Machines

Abstract: A system’s behavior is typically specified through models such as state diagrams that describe how the system should behave. According to researchers, it is not clear what a state actually represents regarding the system to be modeled. Standards do not provide adequate definitions of or sufficient guidance on the use of states. Studies show these inconsistencies can lead to poor or incomplete specifications, which in turn could result in project delays or increase the cost of the system design. This paper aims to establish a precise definition of the notion of states and state machines, a goal motivated by system modelers’ (e.g., requirement engineers’) need to understand key concepts and vocabulary such as states and state machine, which are major behavioral modeling tools (e.g., in UML). “State” is the main notion of a state machine in which events drive state changes. This raises questions about the nature of these state-related notations. The semantics of these concepts is based on a new modeling methodology called the thinging machine applied to a number of examples of existing models. The thinging machine semantics is founded on five elementary actions that divide the static model into changes/states upon which events are defined.

Machine Learning Techniques for Software Bug Prediction: A Systematic Review

Abstract: The goal of software bug prediction is to identify the software modules that will have the likelihood to get bugs by using some fundamental project resources before the real testing starts. Due to high cost in correcting the detected bugs, it is advisable to start predicting bugs at the early stage of development instead of at the testing phase. There are many techniques and approaches that can be used to build the prediction models, such as machine learning. This technique is widely used nowadays because it can give accurate results and analysis. Therefore, we decided to perform a review of past literature on software bug prediction and machine learning so that we can understand better about the process of constructing the prediction model. Not only we want to see the machine learning techniques that past researchers used, we also assess the datasets, metrics and performance measures that are used during the development of the models. In this study, we have narrowed down to 31 main studies and six types of machine learning techniques have been identified. Two public datasets are found to be frequently used and object-oriented metrics are the highly chosen metrics for the prediction model. As for the performance measure, both graphical and numerical measures are often used to evaluate the performance of the models. From the results, we conclude that the machine learning technique can predict the bug, but there are not many applications in this area that exist nowadays. There are a few challenges in constructing the prediction model. Thus, more studies need to be carried out so that a well-formed result is obtained. We also provide a recommendation for future research based on the results we got from this study.

An Adaptive Thresholding Algorithm-Based Optical Character Recognition System for Information Extraction in Complex Images

Abstract: Extracting texts from images with complex backgrounds is a major challenge today Many existing Optical Character Recognition (OCR) systems could not handle this problem As reported in the literature, some existing methods that can handle the problem still encounter major difficulties with extracting texts from images with sharp varying contours, touching word and skewed words from scanned documents and images with such complex backgrounds There is, therefore, a need for new methods that could easily and efficiently extract texts from these images with complex backgrounds, which is the primary reason for this work This study collected image data and investigated the processes involved in image processing and the techniques applied for data segmentation It employed an adaptive thresholding algorithm to the selected images to properly segment text characters from the image’s complex background It then used Tesseract, a machine learning product, to extract the text from the image file The images used were coloured images sourced from the internet with different formats like jpg, png, webp and different resolutions A custom adaptive algorithm was applied to the images to unify their complex backgrounds This algorithm leveraged on the Gaussian thresholding algorithm The algorithm differs from the conventional Gaussian algorithm as it dynamically generated the blocksize to apply threshing to the image This ensured that, unlike conventional image segmentation, images were processed area-wise (in pixels) as specified by the algorithm at each instance The system was implemented using Python 36 programming language Experimentation involved fifty different images with complex backgrounds The results showed that the system was able to extract English character-based texts from images with complex backgrounds with 697% word-level accuracy and 819% character-level accuracy The proposed method in this study proved to be more efficient as it outperformed the existing methods in terms of the character level percentage accuracy

Towards a Frame of Reference for Smart City Strategy Development and Governance

Abstract: Despite the importance of the smart city concept, few works address how to define and implement smart cities in a clear manner. Furthermore, the smart city literature provides heterogeneous studies and solutions; this heterogeneity creates misunderstanding regarding the smart city definition and strategy. Moreover, stakeholders have multiple and conflicting interests and concerns, which also increase the ambiguity regarding the smart city concept and approach. To meet this challenge and fill this gap, a smart city frame of reference is needed to frame and guide smart city strategy formulation and implementation. In this perspective, the current research conducts a quantitative analysis of various smart city frameworks and strategies, in order to find and demonstrate the common building blocks of a smart city framework. Based on the quantitative analysis, this work proposes a clear and integrative smart city framework. This framework aims to reduce the misunderstanding and ambiguity regarding smart city definitions and strategies by providing a standard smart city approach that fits all smart city contexts. To this effect, the proposed framework considers all smart city concerns and it is composed of the following blocks: Strategic awareness, business strategic planning, IT investment decisions, IT organizational structure, steering committee, IT prioritization process, IT strategic planning, IT budgeting, marketing plan, IT reaction capacity, IT reporting and management strategy.

Investigate the Extrinsic and Intrinsic Motivational Factors of Knowledge Transfer in the Hospitals

Abstract: The value of services in knowledge-intensive organizations like the hospitals is created by the tacit knowledge of the health staff. The weak knowledge transfer activities among the health staff effects on the performance level of the health care services. Hospitals are dependent on continuously learning from mistakes and to make improvements and then to transfer this newly acquired knowledge between units and workers in the organization. It is important to enhance the knowledge transfer behaviors of knowledge source and recipients in the hospitals. This study aims to investigate the extrinsic and intrinsic motivational factors that would be adopted to improve the knowledge transfer processes among the health staff. The data of this study were collected from 475 doctor and nurses that work in three Jordanian hospitals; Albashir Hospital, Jordan University Hospital and Al-Issra Hospital. The collected data were analyzed using many tests such as validity, reliability, demographic and descriptive analysis. The significant results show that there are many extrinsic motivational factors would be adopted to improve the knowledge transfer by the knowledge source and recipients such as reward, ideal salaries, promotions, satisfaction of work position, stability of labor and reputation feedback. On the other hand, there are many intrinsic motivational factors that would be applied to enhance the knowledge transfer such as arduous relationship, altruism, shared understanding, champions and enjoyment to help others. This study helps the hospitals to identify the motivational factors that would be adopted to improve the knowledge transfer processes in order to enhance the quality of health care services.

Classification Algorithm in Predicting the Diabetes in Early Stages

Abstract: Diabetes is a standout amongst the deadliest and Chronical diseases which can increase the blood sugar in the human body. Diabetes gives several complications if it is not diagnosed and treated where it might lead to lifeless. Diabetes could be actively controlled when it is primary predicted. To solve this problem and to predict the diabetes in early stage, the machine learning process is used. In this research work, the classifiers like Naive Bayes, KSTAR, ZeroR, OneR, J48 and Random Forest are implemented to predict the diabetes at primary point. Diabetes dataset is sourced from UCI repository and used for this study. The results are evaluated against the performance, accuracy and time. This research work shows the Naive Bayes classification algorithms is the best in predicting the diabetes diseases in primary stage where it helps the health professional to start in diagnosing the patient for diabetes and to save the patient life.

Preliminary Analysis of Malware Detection in Opcode Sequences within IoT Environment

Abstract: With the technological development and means of communication, the Internet of Things (IoT) has become an essential role in providing many services in daily life through millions of heterogeneous but interconnected devices and nodes. This development is opening to many security and privacy challenges that can cause complete network breakdown, bypassed access control or the loss of critical data. This paper attempts to provide a preliminary analysis for malware detection within data generated by IoT-based devices and services in the form of operational codes (Opcode) sequences. Three machine learning algorithms are evaluated and compared for accuracy, precision, recall and F-measure. The results showed that the Random Forest (RF) achieved the best accuracy of 98%, followed by SVM and k-NN, both with 91%. The results are further analyzed based on the Receiver Operating Characteristic (ROC) curve and Precision-Recall curve to further illustrate the difference in performance of all three algorithms when dealing with IoT data.

Hybrid Ant Colony Optimization and Genetic Algorithm for Rule Induction

Abstract: In this study, a hybrid rule-based classifier namely, ant colony optimization/genetic algorithm ACO/GA is introduced to improve the classification accuracy of Ant-Miner classifier by using GA. The AntMiner classifier is efficient, useful and commonly used for solving rulebased classification problems in data mining. Ant-Miner, which is an ACO variant, suffers from local optimization problem which affects its performance. In our proposed hybrid ACO/GA algorithm, the ACO is responsible for generating classification rules and the GA improves the classification rules iteratively using the principles of multi-neighborhood structure (i.e., mutation and crossover) procedures to overcome the local optima problem. The performance of the proposed classifier was tested against other existing hybrid ant-mining classification algorithms namely, ACO/SA and ACO/PSO2 using classification accuracy, the number of discovered rules and model complexity. For the experiment, the 10-fold cross-validation procedure was used on 12 benchmark datasets from the University California Irwine machine learning repository. Experimental results show that the proposed hybridization was able to produce impressive results in all evaluation criteria.

Pedagogical Resources Indexation Based on Ontology in Intelligent Recommendation System for Contents Production in d-Learning Environment

Abstract: We witness, today, a strong evolution of learning environments. In parallel, a problem has emerged, consisting in how to capitalize the production of resources when switching from one environment to another. The heterogeneity of the environments, the evolution of the platforms and the will to reuse the educational resources already produced pushed us to design an intelligent system based on cases. In this study, we will focus on the need for resource indexing to facilitate the task of researching and recommending educational resources for authors regardless of the learning environment used. In the literature, this representation can take two forms: Standards or ontologies. The use of standards has partially solved our problem since it is very beneficial for systems that are under construction. On the other hand, it is more interesting to go through the ontologies for systems that are already designed, that we wish to reuse, especially for those that have shown, through the authors, a great satisfaction in the field of knowledge management. Indeed, their use does not require an investment in the environments concerned by the reuse.

A Comparison between Conditional Random Field and Structured Support Vector Machine for Arabic Named Entity Recognition

Abstract: The Named Entity Recognition (NER) is an integrated task in many NLP applications such as machine translation, Information extraction and question answering. Arabic is one of the authorised spoken languages in the united nation. Currently, there is much Arabic information on the internet, so, nowadays the need for tools which process this information becomes significant. In this study, we have examined the impact of the conditional random field and the structured support vector machine in the task of Arabic NER. The structured support vector machine is the first time to be applied in the Arabic name entity recognition. Our proposed system has three stages: Preprocessing, extracting features and building model. We have used simple features like the bag of words in the [-1,1] window, the bag of part of speech tag in the [-1,1] window to enable our system to detect the multi-words entities. Also, we have tried to enhance the Stanford part of speech tagger to enhance the tagger output tags, which enabled our system to differentiate between the name entities from the non-entities. In addition, we have employed the binary features of: Is a person, is a prename, is a pre-location, is a location and is an organization. Our system has been trained and tested on part of ANER Crop. The results have proved that the conditional random field-based Arabic NER system outperforms the structured support vector machine-based Arabic NER using the same features set.

The reliability of routing protocols as an important factor for road safety applications in VANET-based autonomous cars

Abstract: The autonomous cars are considered as a tremendous disruptive innovation in the coming years. They enable a driving automation system to replace human drivers to control the vehicle with better recognition, decision and driving skills and ultimately enhance the road users’ experience and traffic safety. They can communicate with other cars as they are ready with the Vehicle to Vehicle (V2V) communication technology based on Vehicular Ad hoc Networks (VANETs). One of the objectives of V2V communication is for the safety of all road users. Adequate reliability of routing protocol is subject of concern and must be taken into account to reach an immense standard of road safety accurately and timely. Having no reliability the critical road safety messages will be useless; consequently, the accident that might happen is unable to prevent or avoid. The purpose of this research is to investigate and analyze the quantitative measure of reliability. The reliabilities of a reactive single-path AODV and a multi-path AOMDV routing protocols that comply with road safety requirements in various traffic conditions are studied. The traffic conditions that may impact the internetworking of autonomous cars include node density, size of road area and speed of the nodes. The methods used in this study are based on simulations by using Network Simulator version-2 (NS2) as a network simulator and Simulation of Urban Mobility (SUMO) as a mobility simulator. The simulation results show that both routing protocols, a single-path AODV and a multi-path AOMDV, satisfy the road safety requirements in some conditions. AODV is better in packet delivery, whereas AOMDV has a better performance on average end to end delay. This study is expected to contribute to the determination of the appropriate protocol for use in road safety applications under certain traffic conditions. In conclusion, the reliability of routing protocol is an essential factor to consider in the operation of VANET-based autonomous cars so that the safety and comfort of road users can be guaranteed.

Diabetic Retinopathy Detection using Deep Learning Techniques

Abstract: Diabetic Retinopathy is a type of eye condition induced by diabetes, which damages the blood vessels in the retinal region and the area covered with lesions of varying magnitude determines the severity of the disease. It is one of the most leading causes of blindness amongst the employed community. A variety of factors are observed to play a role in a person to get this disease. Stress and prolonged diabetes are two of the most critical factors to top the list. This disease, if not predicted early, can lead to a permanent impairment of vision. If predicted in advance, the rate of impairment can be brought down or averted. However, it is not easy to detect the presence of this disease, given the time-consuming and tedious process of diagnosis. Presently, digital color photographs are evaluated manually by trained clinicians to observe the presence of lesions caused due to vascular abnormalities, which is the major effect of Diabetic Retinopathy. This method, although it is pretty accurate, proves to be costly. The delay brings out the need to automate the diagnosing, which will, in turn, have a significant positive impact on the health sector. In recent times, the adoption of AI in disease diagnosis has ensured promising and reliable results and this serves as the motivation for this journal. The paper employs Deep learning methodologies for automatic detection of Diabetic Retinopathy, resulting in a maximum accuracy of 80%, as compared to traditional Machine learning approaches giving only a maximum accuracy of 48% on the same IRDiR Disease Grading Dataset (413 images with 5 levels of DR-Training set; 103 images with 5 levels of DR-Test set). The data set contains digital fundus images of different levels of Diabetic Retinopathy in discrete frequency distributions.

Mesh Simplification by Curvature-Enhanced Quadratic Error Metrics

Abstract: Polygonal meshes have a significant role in computer graphics, design and manufacturing technology for surface representation and it is often required to reduce their complexity to save memory. An efficient algorithm for detail retaining mesh simplification is proposed; in particular, the method presented is an iterative edge contraction algorithm based on the work of Garland and Heckberts. The original algorithm is improved by enhancing the quadratic error metrics with a penalizing factor based on discrete Gaussian curvature, which is estimated efficiently through the Gauss-Bonnet theorem, to account for the presence of fine details during the edge decimation process. Experimental results show that this new algorithm helps preserve the visually salient features of the model without compromising performance.

Improving Arabic Instant Machine Translation: The Case of Arabic Triangle of Language

Abstract: Recently, instant translator applications would be a very useful applications when traveling especially when one knows little about the language of the country she/he is traveling to. Arabic to English instant translation has not yet been made available by most applications. In this Article, we attempt to provide mimic way of an application for instant Arabic to English translation. The system provides translation for Arabic Triangle of Language (ālmṯlṯātāllġwyh) which includes Arabic three words that are homographs. The process starts by capturing an image for a homograph using a mobile phone camera, after that the captured word is recognized, taking the diacritic markers into consideration, using an Arabic Optical Character Recognition (OCR). Finally, the system provides an English translation to the homograph. The researchers made use of Histograms of Oriented Gradients (HOG) features and a set of structural and geometrical features of Arabic word segmented and the SVM (multi class) classifier for classification, then providing the English meaning.

Farmer-Friendly Mobile Application for Automated Leaf Disease Detection of Real-Time Augmented Data Set using Convolution Neural Networks

Abstract: In farming, crops are prone to a wide variety of diseases. The impact of sudden climatic change has adverse effects on their growth, providing incubation to harmful viruses and bacteria. Diseases to crops imply a significant negative impact on health, economy and livelihood of the human population. According to the data from the Food and Agricultural Organization (FAO), an average of 1.3 billion tonnes of food crops succumb to such diseases annually. This paper presents an approach to prevent such diseases from propagating, by early diagnosis of such abnormalities in leaves using state of the art deep learning techniques. Convolution Neural Networks based model with additional pre-processing techniques are developed to classify the leaves into affected and healthy category. Various Deep Learning architectures and hyperparameter tuning were carried out and the resulting model produces an accuracy of up to 95% with 400 actual leaf images and up to 98% with 3600 augmented datasets. The models are trained on real-life leaf images of crops, captured from an actual agricultural field. A user-intuitive IoT Web Application is developed to capture, process and display the predicted result (disease status) from the model.

Tree Rotations for Dependency Trees: Converting the Head-Directionality of Noun Phrases

Abstract: Corresponding Author: Ika Alfina Faculty of Computer Science, Universitas Indonesia, Depok, Indonesia Email: ika.alfina@cs.ui.ac.id Abstract: To overcome the lack of NLP resources for the low-resource languages, we can utilize tools that are already available for other highresource languages and then modify the output to conform to the target language. In this study, we proposed an approach to convert an Indonesian constituency treebank to a dependency treebank by utilizing an English NLP tool (Stanford CoreNLP) to create the initial dependency treebank. Some annotations in this initial treebank did not conform to Indonesian grammar, especially noun phrases’ head-directionality. Noun phrases in English usually have head-final direction, while in Indonesian is the opposite, head-initial. We proposed a variant of tree rotations algorithm named headSwap for dependency trees. We used this algorithm to convert the head-directionality for noun phrases that were initially labeled as a compound. Moreover, we also proposed a set of rules to rename the dependency relation labels to conform to the recent guidelines. To evaluate our proposed method, we created a gold standard of 2,846 tokens that were annotated manually. Experiment results showed that our proposed method improved the Unlabeled Attachment Score (UAS) with a margin of 32.5% from 61.6 to 94.1% and the Labeled Attachment Score (LAS) with a margin of 41% from 44.1 to 85.1%. Finally, we created a new Indonesian dependency treebank that converted automatically using our proposed method that consists of 25,416 tokens. The dependency parser model built using this treebank has UAS of 75.90% and LAS of 70.38%.

Email Spam Classification Using Gated Recurrent Unit and Long Short-Term Memory

Abstract: High numbers of spam emails have led to an increase in email triage, causing losses amounting to USD 355 million per year. One way to reduce this loss is to classify spam email into categories including fraud or promotions made by unwanted parties. The initial development of spam email classification was based on simple methods such as word filters. Now, more complex methods have emerged such as sentence modeling using machine learning. Some of the most well-known methods for dealing with the problem of text classification are networks with Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU). This study focuses on the classification of spam emails, so both the LTSM and GRU methods were used. The results of this study show that, under the scenario without dropout, the LSTM and GRU obtained the same accuracy value of 0.990183, superior to XGBoost, the base model. Meanwhile, in the dropout scenario, LSTM outperformed GRU and XGboost with each obtaining an accuracy of 98.60%, 98.58% and 98.52%, respectively. The GRU recall score was better than that of LSTM and XGBoost in the scenario with dropouts, each obtaining values of 98.98%, 98.92% and 98.15% respectively. In the scenario without dropouts, LSTM was superior to GRU and XGBoost, with each obtaining values of 98.39%, 98.39% and 98.15% respectively.

Future Emoji Entry Prediction Using Neural Networks

Abstract: In today’s world, textual data has made momentous progress in social media. The rise of digital communication via text has paved the way to emoji, a pictographically represented way of expressing emotions. In digital communication, Emoji gives a visual appeal to the text, which improves communication and new vistas of exchange and creativity. While emoji entry prediction based on text is well optimized, based on the neural network model, predicting the future emojis from images is not so easy due to lack of knowledge on the same. While effective models already exist for generating text descriptions of images, less attention has been given to models of symbolic description. We have used two models for predicting emoji from images, convolutional neural network architecture for image classification and an emoji2vec embedding into word2vec model. We have also done a sentiment analysis of the text for predicting future emoji labels. Our model captures the relation between emojis in an optimized way. This model has optimized the search time for future emoji entry predictions from images.

Intelligent Attendance System Using Artificial Neural Network Based on Students’ Background

Abstract: Determining the rate of student attendance is an important task in determining the completion of the courses. Despite the success of the technology, it is unfortunate that in many academic institutions, the current systems used to detect student absences. Furthermore, one of the crucial problems in the attendance system does not count student background for continuing in the courses. In this study, we propose an intelligent approach for calculating student attendance based on their Grade Point Average (GPA) and their activities, this approach uses Artificial Neural Network (ANN) for proposing an intelligent attendance system to calculate the attendance rating accurately, meaning the system provide a new rating for each student based on their background. The aim of this research is developing an attendance system for motivation students taking attendance or taking high grade in the class. The result of this approach helps the instructor to allow students who have more activities with more absents to continue in the courses, if not the students have low activity should taking high attendance. This system will more efficient for monitoring students in the classes and replacing absent to activity.

Enhancement of GPS Position Accuracy Using Machine Vision and Deep Learning Techniques

Abstract: The accuracy of GPS position estimation in urban cities is an issue which need to be resolved using machine vision and deep learning techniques. The accuracy of GPS in horizontal direction is better than in the vertical direction. Although for most of the navigation applications in intelligent transportation systems, horizontal positioning accuracy is vital, but vertical position accuracy gives idea about road slanting conditions. Several statistical methods like median filtering, homomorphic filtering and k-means clustering, etc., can be used to improve upon the position accuracy of GPS signals. Such methods are useful for offline applications where a lot many GPS measurements are taken at a single point and afterwards filtering is applied to batch of measurement. In this study, the GPS positioning errors which are caused by sensor noise, ionospheric effects, occlusions by building facades, etc., have been considered for online improvement in position estimation using computer vision and deep learning methods by empirically choosing hyper-parameters.

Covid-19 global spread analyzer: An ML-based attempt

Abstract: The novel Coronavirus 2019 (COVID-19) has caused a pandemic disease over 200 countries, influencing billions of humans In this consequence, it is very much essential to the identify factors that correlate with the spread of this virus The detection of coronavirus spread factors open up new challenges to the research community Artificial Intelligence (AI) driven methods can be useful to predict the parameters, risks and effects of such an epidemic Such predictions can be helpful to control and prevent the spread of such diseases In this study, we introduce two datasets, each of which consists of 25 country-level factors and covers 137 countries summarizing different domains COVID-19STC aims to detect the increase of the total cases, whereas COVID-19STD aimed for total death detection For each data set, we applied three feature selection algorithms (vis correlation coefficient, information gain and gain ratio) We also apply feature selection by the Wrapper methods using four classifiers, namely, NaiveBayes, SMO, J48 and Random Forest The GDP, GDP Per Capital, E-Government Index and Smoking Habit factors found to be the main factors for the total cases detection with accuracy of 73% using the J48 classifier The GDP and E-Government Index are found to be the main factors for total deaths detection with accuracy of 71% using J48 classifier © 2020 Rana Husni Al Mahmoud, Eman Omar, Khaled Taha, Mahmoud Al-Sharif and Abdullah Aref