Support Vector Machine and Artificial Neural Networks for Hydrological Cycles Classifications of a Water Reservoir in the Amazon

This paper presents a methodology based on computational intelligence techniques for classification of hydrological cycles that can infer the change in the physico-chemieal parameters and metals from the water of a reservoir in the Amazon. The methodology initially consists in perform a pre-processing the data to select the most relevant variables of the samples. After that, we compared two different machine learning classifiers, namely SVM （support vector machine） and ANN （artificial neural network）. The automatic model selection is made to choose the parameters of the classifiers. The results indicate that the support vector machine classifier using radial basis function or polynomial kernel exhibited superior results to ANN in terms of overall accuracy and robustness. The SVM classifier accuracy （89.1%） can be considered satisfactory, since there is a great variability of physico-chemical parameters and metals in the hydrological cycles and in the different ecosystems where are the sampling station.     

An Ontological Approach to Intelligence Gathering Using Semantic Metadata

Effectively presenting next to real-time knowledgeable answers to runtime user generated queries, from disparate sources requires the implementation of semantics to safeguard the future of knowledge discovery and reuse. The paper explores the foundations of the research for the design of an information gathering tool based on the intelligence domain; focusing on the exploration of semantic metadata, incorporating ontologies and implementing swarm intelligence theory in the conceptualisation of the system design for the IGUSMON Project, in order to present an efficient and innovative solution.     

Performance Evaluation of Lateration,KNN and Artificial Neural Networks Techniques Applied to Real Indoor and Outdoor Location in WSN

WSN （wireless sensors networks） is a promising research area which has been in center of attention in many applications on telecommunications. Despite of many existing applications, a bunch of algorithms have been created or applied to solve different issues surrounding WSN. An attractive subtopic on research area is Localization Algorithms. Due to the countless applications, it is almost impossible to list all the algorithms applicable to solve sensor＇s location problem in WSN over distinguished parameters associated to diverse environments. This paper evaluates a geometrical algorithm, an instance based algorithm and a function approximator algorithm, having the RSS/ （received signal strength indicator） as metric to estimate planar coordinates in an indoor and outdoor environment using a WSN based on IRIS mote. The analysis of the WSN is constructed over statistical data obtained from empirical experiments and the observed characteristics of the algorithms. We also estimate the performance for different parameters configurations applied to the algorithms for both indoor and outdoor environment. Also, some comments about the tradeoff between the accuracy and the performance of the algorithms are made objecting to. Also, some objections about the tradeoff between the accuracy and the performance of the algorithms are made when relevant.     

Using Genetic Algorithm to Support Artificial Neural Network for Intrusion Detection System

The following methods detect the attacks intrusion detection system： ANN （artificial neural network） for recognition and GA （genetic algorithm） for optimization of ANN results. We use KDD-CUP dataset to obtain the results, which shows around 0.9998 accuracy of applied methods in detecting the threads. ANN with GA requires 18 features.     

Study on Robot Path Planning Based on an Improved Artificial Potential Field Method

The method of artificial potential field has obvious advantages among the robot path planning methods including simple structure, small amount of calculation and relatively mature in theory. This paper puts forward the ＂Integral method＂ focusing on solving the problem of local minimization. The method analyses the distribution of obstructions in a given environment and regards adjacent obstacles as a whole, By changing the parameters of the repulsive force field, robots can quickly get out of the minimum point and move to the target point. This paper uses the Simurosot platform to carry on the simulation experiment on the improved artificial potential field method, which projects a feasible path successfully and verifies this method.     

Reference direction based immune clone algorithm for many-objective optimization

In this paper, a new preference multi-objective optimization algorithm called immune clone algorithm based on reference direction method （RD-ICA） is proposed for solving many-objective optimization problems. First, an intelligent recombination operator, which performs well on the functions comprising many parameters, is introduced into an immune clone algorithm so as to explore the potentially excellent gene segments of all individuals in the antibody pop- ulation. Second, a reference direction method, a very strict ranking based on the desire of decision makers （DMs）, is used to guide selection and clone of the active population. Then a light beam search （LBS） is borrowed to pick out a small set of individuals filling the external population. The proposed method has been extensively compared with other recently proposed evolutionary multi-objective optimization （EMO） approaches over DTLZ problems with from 4 to 100 objectives. Experimental results indicate RD-ICA can achieve competitive results.     

A Privacy Taxonomy for the Management of Ubiquitous Environments

Pervasive and ubiquitous environments must handle the detection and management of users, devices and services, while guaranteeing the privacy of both the users and the environment itself. Current techniques for handling privacy found in the literature treating the subject in various ways, while concentrating on the device management, communication protocols, user profiles and environmental access. This paper examines a control model for privacy in pervasive environments from the perspective of the environment. A prototype was devised and tested to validate the generic model of privacy which was also used to compare taxonomic concepts in the literature. Moreover, the prototype was devised and tested to validate the generic model of privacy for control and manage various users, devices and environments and so on. The prototype was based on Percontrol （a system for pervasive user management）, which was only intended to identify users using Wi-Fi, and now it is capable of managing temperature, luminosity and other preferences, measured by a WSN （wireless sensor network） embedded to Percontrol, and the data treatment is done by an ANN （artificial neural network）. Results confirmed the viability of device detection with Wi-Fi, Bluetooth and RFID （radio frequency identification） for an increases slight of the latency in registering new devices on the system.     

Electric Power Marketing based on intelligence decision tree and artificial Neural Network

A new method of using ANN in Electric Power Marketing is introduced in this paper. Intelligence decision tree classification algorithm is used in data preprocessing to ensure data optimization, the algorithm about hidden nodes of ANN is improving and ANN computational complexity is reduced. The experiment result indicates that ANN model has good predicting effect in electric power distribution.     

A Channel Model for Underground Sensor Networks in the Mixture of Crude Oil,Water and Soil

Future enhanced oil recovery technology can greatly benefit from the wireless sensor networks to effectively operate in underground oil reservoirs. In such a case, millimeter scale sensor nodes with antennas at the same scale have to be deployed in the confined underground oil reservoir fractures. This necessitates the sensor nodes to be operating in the THz frequency range. In this paper, the propagation based on electromagnetic （EM） waves in the Terahertz band （0.1-120.0 THz） through a crude oil/water mixture and soil medium is analyzed in order to explore its applicability in underground oil reservoir assessments. The developed model evaluates the total path loss and the absorption loss that an EM wave experiences when propagating through the crude oil/water mixture and soil medium. Our results show that sensors can communicate successfully for distances up to 1 cm. Furthermore, we have determined the existence of two transmission bands, in which the path loss is below 100 dB. Among those, the frequency window, which provides the best performance, is determined as 70 THz to 85 THz. Different path and absorption loss schemes are considered, which suggests that the 70 THz to 85 THz band is suitable for sensor communications in a medium of crude oil/water mixture and soil.     

交叉融合的“人工智能+”学科建设探索与实践

目前,人类正迈入智能时代,人工智能与经济、社会的发展深度融合,成为了孕育新一轮科技革命和产业变革的核心推动力量,培养智能社会建设所需要的跨学科人工智能创新创业人才变得极为必要。为了适应智能时代对人才培养的这一迫切需求,高校需要加强智能类学科建设,探索交叉融合的“人工智能+”学科建设新模式,提升人工智能领域学科地位并创新拓展其他学科的发展建设方向。文中提出了“人工智能+”学科建设的交叉融合模式:一方面,从人工智能学科内涵出发,依托计算机科学与技术学科开展人工智能基础学科方向建设,夯实人工智能领域学科的基础;另一方面,结合社会经济产业发展的重点领域方向布局,依托相关领域优势学科发展建设行业领域智能化新学科方向,实现人工智能学科建设与其他领域学科建设的共融共生、相互助力、协同发展。在重庆邮电大学和重庆市的学科建设中,这一模式发挥了显著作用,为人工智能领域的学科建设提供了有益参考。     

国内外人工智能的研究热点对比与前沿挖掘

为了直观地了解人工智能领域发展现状及研究前沿,剖析国内外研究存在的异同点,助力国内人工智能研究.以Web of Science数据库和CNKI数据库的2008—2019年期刊论文为依据,借助Citespace软件对期刊论文进行科学知识图谱绘制和可视化分析.根据客观数据和科学知识图谱发现:2016年后,人工智能领域迎来新...     

泛逻辑学理论——机制主义人工智能理论的逻辑基础

当前，世界各主要大国都把人工智能作为它们的国家战略。人工智能的发展正在快速改变着人类的生活方式和思想观念。在中国，有一小批研究者20多年来一直在基于辩证唯物主义潜心研究具有普适性的人工智能基础理论，包括智能的形成机制、逻辑基础、数学基础、协调机理、矛盾转化等。终于，他们各自建立了机制主义人工智能理论、泛逻辑学理论、因素空间理论、协调学、可拓学、集对分析等。其中，机制主义人工智能理论是基于智能形成机制的通用理论，它能把现有的结构主义、功能主义和行为主义三大流派有机地统一起来，使意识、情感、理智成为三位一体的关系；因素空间理论是机制主义人工智能理论的数学基础；泛逻辑学理论是机制主义人工智能理论的逻辑基础。本文介绍了泛逻辑学理论的基本思想、理论基础和应用方法，阐明它的理论意义和应用价值。特别需要指出的是，在广义概率论基础上建立的命题泛逻辑（包括刚性逻辑和柔性逻辑），可看成一个完整的命题级智能信息处理算子库，库中完整地包含了全部18种柔性信息处理模式（包括16种布尔信息处理模式），可用类型编码<a,b,e>来严格区分，用它可寻找到适合自己的信息处理算子完整簇来使用。在每一个信息处理模式中，各种不确定性的组合状态由不确定性程度属性编码<k,h,β,e>来严格区分，用它可在本信息处理模式的算子完整簇中精确选择具体的算子来使用。这表明柔性信息处理本质上是一把密码锁，它需要专门的密码<a,b,e>+<k,h,β,e>才能正常打开，不能乱点鸳鸯谱。通过只有18种模式，每种模式可以从最大算子连续变化到最小算子，已经证明了没有一个命题算子被遗漏。     

人和人工智能系统的概念形成过程研究

概念的形成是实现人工智能的基础,为研究人工智能系统中概念的形成过程,从人对事物形成概念的过程出发进行了研究。比较人和人工智能系统的概念形成过程得到了如下特点:人的优势在于能自主地确定对象表象和对象功能中的各种特征和划分等,能在对象、描述性定义和功能性定义对应关系不完备情况下通过思维和联想建立概念;人工智能系统的优势在于丰富的对象表象感知能力,对象的各种特征和划分的长期存储、运算和分析能力;而人工智能的概念形成过程存在的缺点基本与人的概念形成过程的优点对应。因此本文认为人工智能的概念形成过程必须关注因素的智能识别、功能的系统实践和人经验知识的有师学习。现有技术在缺乏人经验知识的情况下,人工智能系统不能自主建立概念和知识库,不能实现智能过程。     

Performance Evaluation of Antenna Receive Diversity in SlMO-OFDM System over Rayleigh Fading Channel

This paper represents a comparative performance evaluation of different diversity combining techniques for a SIMO-OFDM （single-input-multiple-output orthogonal frequency division multiplexing） system over Rayleigh fading channel. OFDM is a key technique for achieving high data rates and spectral efficiency requirements for wireless communication systems. But in scattering environment, the system performances are severely degraded by the effects of multipath fading and inter-symbol interference. In wireless communication systems, antenna diversity is an important technique to combat multipath fading in order to improve the system performance and increase the channel capacity. In this paper, the performance of different diversity combining techniques-SC （selection combining）, EGC （equal gain combining） and MRC （maximal ratio combining） has been analyzed and compared in terms of SNR （signal to noise ratio） and BER （bit error rate） probability. The simulation results show that the maximal ratio combining technique provides maximum performance improvement relative to all other combining schemes by maximizing the SNR of SIMO-OFDM system at the combiner output. The analytic expressions of error probability and effective bit energy to noise ratio correlated with BPSK （binary phase shift keying） modulation have been derived and formulated for N-branch SC, EGC and MRC schemes. The BER characteristics for all three combining techniques are simulated in MATLAB （matrix laboratory） tool box for varying bit energy to noise ratio. Our results also derives that SNR can be improved if the number of receiving antenna is increased, which in turn reduces BER over a Rayleigh fading channel.     

Normalized Efficient Routing Protocol for WSN

WSNs （wireless sensor networks） consist of thousands of tiny nodes having the capability of sensing, computation, and wireless communications. Unfortunately these devices are limited energy devices, that is means we must save energy as much as possible, to increase network life time as long as possible. In this paper we introduce NEER--normalized energy efficient routing protocol that increases network life time through switching between AODV protocol that depends on request-reply routing, and MRPC that depends on residual battery in routing.     

From IP to Start-up： Building a Robust Life Science Incubator

An incubator is an organization that supports new ventures to grow and survive during the early stages. Mainly dedicated to information technology, life science may appear to be the next hot spot for incubators. Are there stabilized good practices？ Are the business models in information technology and in life science comparable when it goes to start-ups＇ incubation？ Leveraging our experience as practitioner in this field and using an inductive methodology, this paper tends to propose simple principles to help build robust incubators in life science, and to contribute to disseminate an entrepreneurial approach through an industry still dominated by blue chips.     

Real-time control of human actions using inertial sensors

Our study proposes a new local model to accurately control an avatar using six inertial sensors in real-time.Creating such a system to assist interactive control of a full-body avatar is challenging because control signals from our performance interfaces are usually inadequate to completely determine the whole body movement of human actors.We use a pre-captured motion database to construct a group of local regression models,which are used along with the control signals to synthesize whole body human movement.By synthesizing a variety of human movements based on actors’control in real-time,this study verifies the effectiveness of the proposed system.Compared with the previous models,our proposed model can synthesize more accurate results.Our system is suitable for common use because it is much cheaper than commercial motion capture systems.     

Counterworking the Operation Disturbance of a Radiators Structure due to External Actions

The operation of a complicated radiators structure creating a required field in a given area is disturbed often due to different external actions. As an example, the impact of metallic objects approaching off to an antenna system with minimal irradiation on user＇s body is analyzed. In this case, the zone itself, where the area of a weak field （a dark spot） is created, is generally not approachable for sensors installation. In order to counteract the disturbance of the system operation due to presence of different metallic objects in close proximity to the radiators, two methods based on two information sources are studied. These information sources are antennas driving currents and fields at a predetermined point outside the given area. As an example, the antenna system used in the compensation method for reducing human＇s body irradiation, to which a metal object approaches, is analyzed. The conditions of severe external actions are considered： a great metal body and a small distance from this body to the dark spot. It is shown that under these conditions of severe external actions the second method based on the field values measuring demonstrates a higher efficiency. A block scheme for structure of an automatic adjustment is suggested. The considered problem is a common one, and developers of radio circuits are faced with them constantly.     

Towards estimating computer users＇ mood from interaction behaviour with keyboard and mouse

The purpose of this exploratory research was to study the relationship between the mood of computer users and their use of keyboard and mouse to examine the possibility of creating a generic or individualized mood measure. To examine this, a field study （n = 26） and a controlled study （n = 16） were conducted. In the field study, interaction data and self-reported mood measurements were collected during normal PC use over several days. In the controlled study, participants worked on a programming task while listening to high or low arousing background music. Besides subjective mood measurement, galvanic skin response （GSR） data was also collected. Results found no generic relationship between the interaction data and the mood data. However, the re- suits of the studies found significant average correlations be- tween mood measurement and personalized regression models based on keyboard and mouse interaction data. Together the results suggest that individualized mood prediction is pos- sible from interaction behaviour with keyboard and mouse.