Journal of Intelligent & Fuzzy Systems - Volume 35, issue 3

Authors: Zhang, Sheng Tai | Wang, Fei Fei | Duo, Fan | Zhang, Ju Liang

Article Type: Research Article

Abstract: Sentiment analysis mainly studies the emotional tendencies of texts from grammar, semantic rules and other aspects. The texts from social network are characterized by less words, irregular grammar, data noise and so on, which have increased the difficulty of emotion analysis. In order to improve the performance of machine learning in sentiment analysis, this study proposed the Majority Decision Algorithm to classify the emotional tendentious of the text in WeChat, combined the characteristics of five classifiers and integrated the classification results of five classifiers, eventually the text can be classified in WeChat. Firstly, this study utilized the BlueStacks to …crawl the cache of WeChat Moment developed by Tencent company. Secondly, the cache was processed by Python to get the WeChat dataset. After the Chinese word segmentation, data cleaning and segmentation, the sentiment classification experiment were carried out using different classifiers. Finally, a Majority Decision Algorithm composed of five classifiers was established. It included, Naive Bayes (sklearn), Naive Bayes (SnowNLP), SVM (linear), SVM (RBF) and SGD. Then, the comparison was carried out between the performance of the algorithm and the five classifiers. Results show that the precision rates of the five classifiers are 0.8598, 0.8154, 0.8511, 0.8739 and 0.8678; the recall rates are 0.8544, 0.8482, 0.9380, 0.9226 and 0.9349; F1 scores are 0.8571, 0.8315, 0.8924, 0.8975 and 0.9001, respectively. The algorithm of the Precision rate, Recall rate and F1 score were 0.8804, 0.9349 and 0.9069, respectively, indicating that algorithm in current study significantly improved the performance, which can be effectively applied into the new text form of WeChat Moment. The study can provide theoretical reference for sentiment classification of Chinese text based on machine learning. Show more

Keywords: WeChat, sentiment classification, Majority Decision Algorithm(MDA)

DOI: 10.3233/JIFS-169653

Citation: Journal of Intelligent & Fuzzy Systems, vol. 35, no. 3, pp. 2975-2984, 2018

Authors: Pu, Z. | Wang, W. | Li, Y. | Huang, H.

Article Type: Research Article

Abstract: To overcome disadvantages of conventional Doppler estimation and compensation for non-stationary underwater platforms, a novel algorithm based on joint time-frequency searching is proposed to track and rapidly compensate the time-variant Doppler factor in underwater acoustic OFDM (Orthogonal Frequency Division Multiplexing) communication. The algorithm utilizes the comb pilot’s subcarriers combined with frequency-domain resampling to calculate the channel impulse response under different Doppler factors. Then the correct Doppler factor is achieved and eliminated by seeking the sparest channel impulse response, which is accessed using compressed sensing method. Experiment results obtained by Monte Carlo simulation and sea-trial in shallow-water are provided, demonstrating that …the new algorithm can effectively track and rapidly compensate the time-variant Doppler factor, thus ensuring the stable underwater acoustic OFDM mobile communication for non-stationary underwater platforms. Show more

Keywords: Underwater acoustic communication, orthogonal frequency division multiplexing, time-variant Doppler tracking, Doppler fast compensation

DOI: 10.3233/JIFS-169654

Citation: Journal of Intelligent & Fuzzy Systems, vol. 35, no. 3, pp. 2985-2996, 2018

Authors: Wang, Fei | Shi, Xiaohua | Lu, Xinghai

Article Type: Research Article

Abstract: Android systems typically run on resource-constrained hand-hold devices. How to efficiently utilize Java heaps is one of the most important issues of concerns to the developers. Developers often use profilers to observe the utilization efficiency of Java objects, hoping to find out memory allocation bottlenecks, identify and solve problems such as memory leaks, etc. However, currently there lacks a low-overhead and efficient Java object profiler on Android and its Java virtual machines. In this paper, we design and implement a novel and low-overhead Java object profiler based on the Address-Chain technique, on Android 6.0 and its ART virtual machine, …which uses an AOT (ahead-of-time) compiler and has complex garbage collection algorithms. Our profiler records the allocation site, the class information of the object, the object size, the birth time and death time of the object, the physical memory trace of the object movements with time stamps, the last access time and the access regular pattern, etc., for every Java object. The data profiled can help the developers to detect memory leaks, implement optimizations like pretenuring and tune the performance of garbage collector, etc. The Java object profiling mechanism proposed in this paper has low execution time overhead, imposes no overhead on the Java heap and does not modify any existing key data structure of the ART Virtual Machine, including the object layouts, class layouts and any others. By caching object access event in global register and removing redundant instrumentation, on Nexus7 and Android 6.0, the read/write barriers overheads of the profiler are about 19% on average for EEMBC, SciMark and other workloads. The I/O overheads are about 28% and the total execution overheads are about 51% on average. Show more

Keywords: Low-overhead profiler, android, managed languages, object lifetime information

DOI: 10.3233/JIFS-169655

Citation: Journal of Intelligent & Fuzzy Systems, vol. 35, no. 3, pp. 2997-3009, 2018

Authors: Wang, Peng | Cheng, Jun | Feng, Wei

Article Type: Research Article

Abstract: Path planning for human beings is based on scene understanding and semantic map as well as geometry features. The point cloud map can help robots to perform path planning. However, it is quite different from the natural way of human beings. In this paper, we propose to construct a novel framework for topological semantic map. Specifically, we construct a 2D semantic map by projecting 3D scene semantic information recognized by convolutional neural network onto a 2D plane. 3D reconstruction of the environment is achieved by RGB-D SLAM 3D space mapping algorithm. The intersections in the 2D map are recognized offline, …and the semantic annotation of the intersection in the topological map is utilized to build up a complete object-based semantic map. Experimental results demonstrate the feasibility of the proposed approach. Show more

Keywords: Topological semantic map, 3D information, convolution neural network, schematic diagram

DOI: 10.3233/JIFS-169656

Citation: Journal of Intelligent & Fuzzy Systems, vol. 35, no. 3, pp. 3011-3020, 2018

Authors: Zhao, Hailei | Jin, Dehuan

Article Type: Research Article

Abstract: In the trading activities, the liquidity level of assets and its impact on future earnings has always been a concern for investors. This paper calculates the liquidity of the stock market by constructing a frictionless asset model. At the same time, we test the application of the liquidity level premium index in the LA-CAPM model, mainly draws the following conclusions: (1) Although the individual liquidity of the stock has different individual characteristics, but showing a common trend of change. Stock individual liquidity level premium level and the stock market performance of the opposite, when the market is good, the stock …individual liquidity level premium is low. (2) The liquidity premium of the market portfolio is generally inversely related to the market index, and the liquidity premium between the different market combinations is significantly different. (3) In different portfolios, the portfolio of large, value-based company stocks is superior to the portfolio of small, growing corporate stocks. The results of this study have important reference significance for the construction of liquidity index and the research of liquidity and asset pricing. Show more

Keywords: Information asymmetry, LA-CAPM Model, stock market, liquidity level, premium measurement

DOI: 10.3233/JIFS-169657

Citation: Journal of Intelligent & Fuzzy Systems, vol. 35, no. 3, pp. 3021-3034, 2018

Authors: Xu, Bing | Liu, Liqun | Zhang, Xiaoping | Wu, Xianjun

Article Type: Research Article

Abstract: In order to effectively improve the data transmission efficiency of wireless sensor networks (WSN), a new wireless sensor network chain routing algorithm is proposed based on adaptive backoff-adjusted medium access control. At first, the optimal transmission back-off time of undetermined frames is estimated by the number of active nodes in this algorithm, and the performance evaluation index of WSN is presented with area coverage, transmission distance and residual energy. At the same time, the number of collisions and transmission attempt rate are given with adaptive backoff-adjusted medium access control, and the chain routing algorithm is build. Finally, the key factors …influencing the algorithm are deeply studied through experimental platform and numerical simulation with MATLAB. The results show that, compared with the traditional chain routing algorithm and progressive best backoff algorithm, the algorithm has great advantages in terms of chaining efficiency, node life cycle and delay. Show more

Keywords: Wireless sensor network, chain routing, energy, chaining, life cycle

DOI: 10.3233/JIFS-169658

Citation: Journal of Intelligent & Fuzzy Systems, vol. 35, no. 3, pp. 3035-3041, 2018

Authors: Zhai, Zhonghua

Article Type: Research Article

Abstract: Generative Adversarial Networks have demonstrated potential on a variety of generative tasks, although they are regarded as unstable and sometimes they miss modes. We propose Auto-encoder Generative Adversarial Networks - a convolutional neural network combining auto-encoders with Generative Adversarial Networks. The former brings more information to Generative Adversarial Networks to reduce problems of miss modes and the latter makes the picture generated more coherent because it can better handle multiple modes in the output. We also show that image composition is available for Auto-encoder Generative Adversarial Networks so that it can be used for many feature-based tasks. Besides, we can …generate different samples by adding a random noise to a feature vector. Show more

Keywords: Generative adversarial networks, auto encoder, joint loss, image composition

DOI: 10.3233/JIFS-169659

Citation: Journal of Intelligent & Fuzzy Systems, vol. 35, no. 3, pp. 3043-3049, 2018

Authors: Tian, Xueying

Article Type: Research Article

Abstract: The proper integration of the fuzzy clustering and neural network will help to enhance the accuracy of recommendation system. In this paper, the author designed an intelligent recommendation system and analyzed its application in smart old-age care. In the content-based recommendation system, we describe frequently the product content with some key words, and this could also be used in the home-based care services. by using chaotic neural network model, we analyze the user behavior and the recommended quality. The results show that the large data can provide accurate service supply for the elderly with different pension modes at different levels …of pension services. At the same time, through the online platform, the intelligent computing and decision-making function of the aged service can be realized. Show more

Keywords: Intelligent recommendation, user behaviour mining, smart old-age care, o2o model, big data

DOI: 10.3233/JIFS-169660

Citation: Journal of Intelligent & Fuzzy Systems, vol. 35, no. 3, pp. 3051-3063, 2018

Authors: Wang, Zhihui

Article Type: Research Article

Abstract: The spectrum cluster algorithm is opposite in the other cluster algorithm has the obvious superiority that can distinguish the non-raised distribution the cluster, suits extremely in many actual problems. With this prior model this paper combines the fuzzy set system to propose the new data mining algorithm. At the same time, we build a fuzzy comprehensive evaluation model and analyze the evaluation system of physical education. The result shows that the fuzzy comprehensive evaluation method is improved by using the grey relational grade, which avoids the disadvantages of adopting the principle of maximum membership degree. Therefore, this evaluation method can …be used in the evaluation of public sports teaching in colleges and universities. Show more

Keywords: Mining algorithm, grey system, public physical education, high dimensional data, fuzzy comprehensive evaluation

DOI: 10.3233/JIFS-169661

Citation: Journal of Intelligent & Fuzzy Systems, vol. 35, no. 3, pp. 3065-3076, 2018

Authors: Huang, Zhen | Shang, Chao | Chen, Yanzhang

Article Type: Research Article

Abstract: In this paper, the spatial dobbin model is used to analyze the spatial impact of local government competition and land price on foreign direct investment. The data came from 69 prefecture level cities in five provinces in Central China, time from 2005–2015. The empirical research shows that the higher the level of regional economic development, the better the road traffic conditions, the lower the industrial land price and the higher the degree of government competition, the more the FDI will flow into the region. The impact of environmental regulation on FDI is not obvious. In other areas, the level of …economic development, the increase of industrial land price and the enhancement of environmental regulation intensity will promote the level of foreign investment in this area, while the traffic improvement in other areas and the intensification of government competition will inhibit FDI in the region. The results show that the means of attracting FDI among governments in order to achieve economic goals are available and effective. Show more

Keywords: Copula function, neural network, government competition, foreign direct investment, space spillover effect

DOI: 10.3233/JIFS-169662

Citation: Journal of Intelligent & Fuzzy Systems, vol. 35, no. 3, pp. 3077-3089, 2018