Measure of the Content Creation Score on Social Network Using Sentiment Score and Passion Point

Social network is one of efficient tools for spreading information. The evaluation of the content creation of a user is a useful feature to improve the ability of information propagation on social network. In this paper, the measures for evaluating the user’s content creation are proposed. They include the passion point of a user with a brand and the quality of the user’s posts. The passion point is computed based on the sentiment score of posting and the activity of the user. The quality of the user’s posts is computed through the analyzing of the post’s content. Those measures are combined to analyze the interesting of posts. The proposed method has been tested and get the positive experimental results.

An Efficient Framework for Vietnamese Sentiment Classification

With the booming development of E-commerce platforms in many counties, there is a massive amount of customers’ review data in different products and services. Understanding customers’ feedbacks in both current and new products can give online retailers the possibility to improve the product quality, meet customers’ expectations, and increase the corresponding revenue. In this paper, we investigate the Vietnamese sentiment classification problem on two datasets containing Vietnamese customers’ reviews. We propose eight different approaches, including Bi-LSTM, Bi-LSTM + Attention, Bi-GRU, Bi-GRU + Attention, Recurrent CNN, Residual CNN, Transformer, and PhoBERT, and conduct all experiments on two datasets, AIVIVN 2019 and our dataset self-collected from multiple Vietnamese e-commerce websites. The experimental results show that all our proposed methods outperform the winning solution of the competition “AIVIVN 2019 Sentiment Champion” with a significant margin. Especially, Recurrent CNN has the best performance in comparison with other algorithms in terms of both AUC (98.48%) and F1-score (93.42%) in this competition dataset and also surpasses other techniques in our dataset collected. Finally, we aim to publish our codes, and these two data-sets later to contribute to the current research community related to the field of sentiment analysis.

Cohesive Subgraph Models for Overlapping Community Search over Networks

There has been significant interest in the study of the problem of community search in large networks. Given one or more query nodes, this problem aims to discover densely connected subgroups containing these nodes. Various algorithms have been proposed to solve this challenging problem using different measures or a variety of cohesive subgraphs. In this paper, given an undirected graph and a set of query nodes, we study the community search using novel several cohesive subgraph models. More precisely, we propose to exploit several cohesive structures in a unified framework to find densely communities for query nodes in large complex networks. First, we review some existing cohesive structures. Next, to make these structures more effective models of communities, we focus on interesting configurations that are larger and more cohesive by fulfilling some constraints. The new structures obtained allow to ensure a larger density on the discovered communities and overcome some weaknesses of existing models. Finally, empirical results show the effectiveness of our framework to find communities for query nodes in a variety of real graphs.

Non-Attractive Periodic Trajectory Formation Mechanism on Random and Chaotic Time Series

Pseudo-random number series extracted from chaotic and random time series from the chaotic and random neural network (CRNN) with fixed-point arithmetic has been the focus of attention for protecting the information security of IoT devices. Pseudo-random number series generated by a computer is eventually periodic, practically. The produced closed trajectory is not a limit cycle, because which does not divide the phase space into 2 regions. The closed trajectory in this work is called a non-attractive periodic trajectory (NPT) because it hardly attracts trajectories within the neighborhood. The method of preventing the closed trajectory formation has been proposed on the basis of the NPT formation mechanism in this paper. The method has extended the period of NPT considerably. It is expected to apply security applications for IoT devices.

A Method for Image Forgery Detection Based on Error Level Analysis (ELA) Technique

In the last years, there has been growing a large increase in digital imaging techniques, and their applications became more and more pivotal in many critical scenarios. Conversely, hand in hand with this technological boost, imaging forgeries have increased more and more along with their level of precision. In this view, the use of digital tools, aiming to verify the integrity of a certain image, is essential. Indeed, insurance is a field that extensively uses images for filling claim requests and a robust forgery detection is essential. This paper proposes an approach which aims to introduce a full-automated system for identifying potential splicing frauds in images of car plates by overcoming traditional problems using artificial neural networks (ANN). For instance, classic fraud-detection algorithms are impossible to fully automatize whereas modern deep learning approaches require vast training datasets that are not available most of the time. The method developed in this paper uses Error Level Analysis (ELA) performed on car license plates as an input for a trained model which is able to classify license plates in either original or forged.

Intelligent Content Driving of Engineering Model System in Modeling Platform

Engineering modeling software systems have been developed during a long integration process from separated partial solutions to current modeling software platforms (MSPs). MSP is expected to provide all necessary model creation and application capabilities during integrated innovation and the life cycle of commercial and industrial products (CIP). Recently, advanced CIP is operated by component systems organized within an increasingly autonomous cyber physical system (CPS). CIP is represented by the engineering model system (EMS). EMS is driven by active contexts between the outside world and EMS, between component models of EMS, and between objects in a component model. EMS reacts to any new contribution using all formerly represented contexts. Consistent structure of contexts gives autonomous operation capability for EMS. Active contexts between the outside world and EMS make EMS sensitive to outside world changes. In the other direction, EMS can generate advice for the outside world using high level and well-organized active knowledge as context. Contributing to research in key issues around EMS and the relevant software technology, this paper introduces results in requirements against MSP capabilities to represent intelligent driving content (IDC) in EMS. A novel organized structure of IDC and continuous engineering (CE) aspects of IDC development are explained and discussed placing the main emphasis on situation awareness. Finally, a new concept is introduced in which purposeful EMS acts as the only media in communication of researchers. Specially configured MSP facilitates participation from industrial, institutional, and academic organizations. The research proceeds at the Laboratory of Intelligent Engineering Systems (IESL) in the organization of the Óbuda University.

The Best Ensemble Learner of Bagged Tree Algorithm for Student Performance Prediction

Student performance is the most factor that can be beneficial for many parties, including students, parents, instructors, and administrators. Early prediction is needed to give the early monitor by the responsible person in charge of developing a better person for the nation. In this paper, the improvement of Bagged Tree to predict student performance based on four main classes, which are distinction, pass, fail, and withdrawn. The accuracy is used as an evaluation parameter for this prediction technique. The Bagged Tree with the addition of Bag, AdaBoost, RUSBoost learners helps to predict the student performance with the massive datasets. The use of the RUSBoost algorithm proved that it is very suitable for the imbalance datasets as the accuracy is 98.6% after implementing the feature selection and 99.1% without feature selection compared to other learner types even though the data is more than 30,000 datasets.

Oversampling Based on Data Augmentation in Convolutional Neural Network for Silicon Wafer Defect Classification

Silicon wafer defect data collected from fabrication facilities is intrinsically imbalanced because of the variable frequencies of defect types. Frequently occurring types will have more influence on the classification predictions if a model gets trained on such skewed data. A fair classifier for such imbalanced data requires a mechanism to deal with type imbalance in order to avoid biased results. This study has proposed a convolutional neural network for wafer map defect classification, employing oversampling as an imbalance addressing technique. To have an equal participation of all classes in the classifier’s training, data augmentation has been employed, generating more samples in minor classes. The proposed deep learning method has been evaluated on a real wafer map defect dataset and its classification results on the test set returned a 97.91% accuracy. The results were compared with another deep learning based auto-encoder model demonstrating the proposed method, a potential approach for silicon wafer defect classification that needs to be investigated further for its robustness.

REST API Auto Generation: A Model-Based Approach

Most of software products, especially mobile applications (apps) rely on a back-end web services to communicate with a shared data repository. Statistics have demonstrated exponential demand on web services, mainly REST, due to the continuous adoption of IoT (Internet of Things) and Cloud Computing. However, the development of back-end REST web services is not a trivial task, and can be intimidating even for seasoned developers. Despite the fact that there are several studies that focus on automatic generation of REST APIs, we argue that those approaches violate the rules of code flexibility and are not appropriate for novice developers. In this study, we present an approach and a framework, named RAAG (REST Api Auto-Generation), that aims to improve productivity by simplifying the development of REST web services. Our RAAG framework abstracts layers, where code generation has been avoided due its limitations. A preliminary evaluation shows that RAAG can significantly improves development productivity and is easy to operate even by novice developers.

Keyphrase Graph in Text Representation for Document Similarity Measurement

To represent the text document more expressively, a kind of graph-based semantic model is proposed, in which more semantic information among keyphrases as well as the structural information of the text are incorporated. The method produces structured representations of texts by utilizing common, popular knowledge bases (e.g. DBpedia, Wikipedia) to acquire fine-grained information about concepts, entities, and their semantic relations, thus resulting in a knowledge-rich interpretation. We demonstrate the benefits of these representations in the task of document similarity measurement. Relevance evaluation between two documents is done by calculating the semantic similarity between two keyphrase graphs that represent them. Experimental results show that our approach outperforms standard baselines based on traditional document representations, and able to come close in performance to the specialized methods particularly tuned to this task on the specific dataset.