Kuan-Cheng Lin

Feature selection based on an improved cat swarm optimization algorithm for big data classification

Feature selection, which is a type of optimization problem, is generally achieved by combining an optimization algorithm with a classifier. Genetic algorithms and particle swarm optimization (PSO) are two commonly used optimal algorithms. Recently, cat swarm optimization (CSO) has been proposed and demonstrated to outperform PSO. However, CSO is limited by long computation times. In this paper, we modify CSO to present an improved algorithm, ICSO. We then apply the ICSO algorithm to select features in a text classification experiment for big data. Results show that the proposed ICSO outperforms traditional CSO. For big data classification, the results show that using term frequency-inverse document frequency (TF-IDF) with ICSO for feature selection is more accurate than using TF-IDF alone.

Feature selection and parameter optimization of support vector machines based on modified cat swarm optimization

Recently, applications of Internet of Things create enormous volumes of data, which are available for classification and prediction. Classification of big data needs an effective and efficient metaheuristic search algorithm to find the optimal feature subset. Cat swarm optimization (CSO) is a novel metaheuristic for evolutionary optimization algorithms based on swarm intelligence. CSO imitates the behavior of cats through two submodes: seeking and tracing. Previous studies have indicated that CSO algorithms outperform other well-known metaheuristics, such as genetic algorithms and particle swarm optimization. This study presents a modified version of cat swarm optimization (MCSO), capable of improving search efficiency within the problem space. The basic CSO algorithm was integrated with a local search procedure as well as the feature selection and parameter optimization of support vector machines (SVMs). Experiment results demonstrate the superiority of MCSO in classification accuracy using subsets with fewer features for given UCI datasets, compared to the original CSO algorithm. Moreover, experiment results show the fittest CSO parameters and MCSO take less training time to obtain results of higher accuracy than original CSO. Therefore, MCSO is suitable for real-world applications.

Feature Selection and Parameter Optimization of Support Vector Machines Based on Modified Artificial Fish Swarm Algorithms

Rapid advances in information and communication technology have made ubiquitous computing and the Internet of Things popular and practicable. These applications create enormous volumes of data, which are available for analysis and classification as an aid to decision-making. Among the classification methods used to deal with big data, feature selection has proven particularly effective. One common approach involves searching through a subset of the features that are the most relevant to the topic or represent the most accurate description of the dataset. Unfortunately, searching through this kind of subset is a combinatorial problem that can be very time consuming. Meaheuristic algorithms are commonly used to facilitate the selection of features. The artificial fish swarm algorithm (AFSA) employs the intelligence underlying fish swarming behavior as a means to overcome optimization of combinatorial problems. AFSA has proven highly successful in a diversity of applications; however, there remain shortcomings, such as the likelihood of falling into a local optimum and a lack of multiplicity. This study proposes a modified AFSA (MAFSA) to improve feature selection and parameter optimization for support vector machine classifiers. Experiment results demonstrate the superiority of MAFSA in classification accuracy using subsets with fewer features for given UCI datasets, compared to the original FASA.

Data mining in emotion color with affective computing

This research applies an innovative way to measure and identify user’s emotion with different ingredient color. How to find an intuitive way to understand human emotion is the key point in this research. The RGB color system that is widely used of all forms computer system is an accumulative color system in which red, green, and blue light are added together showing entire color. This study was based on Thayer’s emotion model which classifies the emotions with two vectors, valence and arousal, and gathers the emotion color with RGB as input for calculating and forecasting user’s emotion. In this experiment, using 320 data divide to quarter into emotion groups to train the weight in the neural network and uses 160 data to prove the accuracy. The result reveals that this model can be valid reckon the emotion by reply color response from user. In other hand, this experiment found that trend of the different ingredient of color on Cartesian coordinate system figures out the distinguishing intensity in RGB color system. Via the foregoing detect emotion model is going to design an affective computing intelligence framework try to embed the emotion component in it.

Web-Based Appreciation and Peer-Assessment for Visual-Art Education

This study describes the application of a Web-based portfolio for appreciation and peer assessment for visual-art education in elementary school. Besides examining the effectiveness of the proposed system in enhancing visual art education, this study also addresses how Web-based portfolios can help teachers to assess student learning progress and facilitate peer assessment. The results of questionnaires and interviews shows that the proposed portfolio system can help student learning and the peer assessment component of the proposed portfolio system can help the students’ learning in visual-art education in elementary school.

Feature Selection for Support Vector Machines Base on Modified Artificial Fish Swarm Algorithm

Feature selection is a search process to find the optimal feature subset to describe the characteristics of dataset exactly. Artificial Fish Swarm Algorithm is a novel meta-heuristic search algorithm, which can solve the problem of optimization by simulate the behaviors of fish swarm. This study proposes a modified version of Artificial Fish Swarm Algorithm to select the optimal feature subset to improve the classification accuracy for Support Vector Machines. The empirical results showed that the performance of the proposed method was superior to that of basic version of Artificial Fish Swarm Algorithm.

Botnet Detection Using Support Vector Machines with Artificial Fish Swarm Algorithm

Because of the advances in Internet technology, the applications of the Internet of Things have become a crucial topic. The number of mobile devices used globally substantially increases daily; therefore, information security concerns are increasingly vital. The botnet virus is a major threat to both personal computers and mobile devices; therefore, a method of botnet feature characterization is proposed in this study. The proposed method is a classified model in which an artificial fish swarm algorithm and a support vector machine are combined. A LAN environment with several computers which has infected by the botnet virus was simulated for testing this model; the packet data of network flow was also collected. The proposed method was used to identify the critical features that determine the pattern of botnet. The experimental results indicated that the method can be used for identifying the essential botnet features and that the performance of the proposed method was superior to that of genetic algorithms.

Mobile Distributed Web Server System

The Internet traffic is growing rapidly in recent years. How to distribute the network traffic effectively is an important issue. In this paper, we propose a mobile distributed web server system to distribute Web content to the mobile servers around the world in inexpensive and effective way. The proposed method is fully compatible with current existing systems. In this paper, we discuss the Web content requesting imbalance (WCRI) problem and describe how this system is able to solve this problem. We also provide a possible implementation of a mobile distributed Web server system.

Augmenting teacher-student interaction in digital learning through affective computing

Interactions between teachers and students can be effectively enhanced if teachers can capture the spontaneous nonverbal behaviors (e.g., facial expressions and body language) of their students in real time, thereby effectively improving teaching strategies and the learning effectiveness of students. In this study, we implemented an expression–response analysis system (ERAS) to analyze facial expressions. The ERAS employs a web camera to capture the facial images of students. Their facial expressions are analyzed to assess their attitude toward progressively more difficult course content, and to determine the relationship between their social interactions and learning effectiveness. The ERAS identified 10 facial feature points that form 11 facial action units (AUs). Subsequently, the AUs were classified as positive, neutral, and negative social interactions by applying a rule-based expert system, and cognitive load theory was applied to verify the classifications. The experimental results showed that student with high coding abilities could adapt to the multimedia digital learning content, as evidenced by the comparatively higher expression of neutral and positive social interactions, whereas students with low coding abilities reported a higher frequency of negative social interactions resulting from the increase in cognitive load. Simultaneously, the real time detection of social interactions can provide a basis for diagnosing student learning difficulties and assist teachers in adjusting their teaching strategies.

Adaptive SVM-Based classification systems based on the improved endocrine-based PSO algorithm

In this study, we proposed an wrapped feature selection and SVMs kernel parameters optimization scheme using Improved Artificial Endocrine System to get an optimal support vector machines classification system. By taking the advantage of the mechanisms of hormone action in Artificial Endocrine System, we can avoid to obtain local optimums and oscillations. We used the UCI database to evaluate the performance of the proposed scheme with the previous methods. The experiment results indicated that the proposed scheme can avoid local optimum and also reduce feature numbers significantly with a good-enough accuracy in high-complexity datasets. Moreover, by decreasing the number of unnecessary features, we can even improve the accuracy of classification.