Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Data Mining - Concepts and Techniques.

How have Japanese companies become world leaders in the automotive and electronics industries, among others? What is the secret of their success? Two leading Japanese business experts, Ikujiro Nonaka and Hirotaka Takeuchi, are the first to tie the success of Japanese companies to their ability to create new knowledge and use it to produce successful products and technologies. In The Knowledge-Creating Company, Nonaka and Takeuchi provide an inside look at how Japanese companies go about creating this new knowledge organizationally. The authors point out that there are two types of knowledge: explicit knowledge, contained in manuals and procedures, and tacit knowledge, learned only by experience, and communicated only indirectly, through metaphor and analogy. U.S. managers focus on explicit knowledge. The Japanese, on the other hand, focus on tacit knowledge. And this, the authors argue, is the key to their success--the Japanese have learned how to transform tacit into explicit knowledge. To explain how this is done--and illuminate Japanese business practices as they do so--the authors range from Greek philosophy to Zen Buddhism, from classical economists to modern management gurus, illustrating the theory of organizational knowledge creation with case studies drawn from such firms as Honda, Canon, Matsushita, NEC, Nissan, 3M, GE, and even the U.S. Marines. For instance, using Matsushita's development of the Home Bakery (the world's first fully automated bread-baking machine for home use), they show how tacit knowledge can be converted to explicit knowledge: when the designers couldn't perfect the dough kneading mechanism, a software programmer apprenticed herself withthe master baker at Osaka International Hotel, gained a tacit understanding of kneading, and then conveyed this information to the engineers. In addition, the authors show that, to create knowledge, the best management style is neither top-down nor bottom-up, but rather what they call "middle-up-down," in which the middle managers form a bridge between the ideals of top management and the chaotic realities of the frontline. As we make the turn into the 21st century, a new society is emerging. Peter Drucker calls it the "knowledge society," one that is drastically different from the "industrial society," and one in which acquiring and applying knowledge will become key competitive factors. Nonaka and Takeuchi go a step further, arguing that creating knowledge will become the key to sustaining a competitive advantage in the future. Because the competitive environment and customer preferences changes constantly, knowledge perishes quickly. With The Knowledge-Creating Company, managers have at their fingertips years of insight from Japanese firms that reveal how to create knowledge continuously, and how to exploit it to make successful new products, services, and systems.

The knowledge-creating company

In recent years, the wireless body area network WBAN has emerged as a new technology for e-healthcare applications. The WBANs promise to revolutionize health monitoring. However, this technology remains in the first stages and much research is underway. Designers of such systems face a number of challenging tasks, as they need to address conflicting requirements. This includes managing the network, the data, while maximizing the autonomy of each network node. Reducing the consumption of a node, the management of network resources and security insurance are therefore major challenges. This paper presents a survey of body area networks including the WBANs challenges and -architecture, the most important body sensor devices, as well as sensor board hardware and platforms. Further, various applications of WBANs in the medical field are discussed, as well as wireless communications standards and technologies. The newest researches related to WBANs at physical and MAC layers are presented. Finally the paper identifies data security and privacy in WBANs as well as open research issues.

A Comprehensive Overview of Wireless Body Area Networks WBAN

The increase in the average age of the world's population over the coming decades will impact several aspects of our lives. In particular, it will increase the demand and challenges of delivering care and assistance to elderly or fragile people. This article describes an ambient assisted living system, called Bridge (behaviour drift compensation for autonomous and independent living), that's based on a set of home services that include a wireless sensor-actuator network for detection of inhabitants' behaviors and a rich, flexible communication system between the inhabitants and a social environment aimed at reassuring both inhabitants and their families. All Bridge solutions have been derived from real needs with the help of a social mediator and tested by users in their living environments.

Bridge: Mutual Reassurance for Autonomous and Independent Living

Defining activity models in order to monitor human behavior in smart environments is one of the major issues at the moment of building systems of activity supervision for diagnosis, prediction and control. For the purpose of addressing this problem, this paper proposes a general theoretical approach based on the use of a Knowledge Engineering methodology and a Machine Learning process, which are funded on a general theory of dynamic process modeling, the Timed Observation Theory.

Levels of Abstraction for Behavior Modeling in the GerHome Project

Cemagref, Unite´ Ouvrages Hydrauliques et Hydrologie, Aix en Provence, Franceemilie.masse@cemagref.fr, corinne.curt@cemagref.frKeywords: Multi modeling, diagnosis reasoning, dynamic systemAbstract: Thispaper presents amodelling approach of dynamic process for diagnosis that iscompatible withthe Stochas-tic Approach framework for discovering temporal knowledge from the timed observations contained in adatabase. The motivation is to deﬁne a multi-model formalism that is able to represent both the knowledgeof these two sources. The aim is to model the process at the same level of abstraction that an expert uses todiagnose the process. The underlying idea is that at this level of abstraction, the model is simple enough toallow an efﬁcient diagnosis. The proposed formalism repres ents the knowledge in four models: a structuralmodel deﬁning the components and the connection relations o f the process, a behavioural model deﬁning thestates and the transitions states of the process, a functional model containing the logical relations between thevalues of the process’s variables, which are deﬁned in the pe rception model. The models are linked with theprocess’s variables. This point facilitates the analysis of the consistency of the four models and is the basis ofthe corresponding knowledge modelling methodology. The formalism and the methodology is illustrated withthe model of a hydraulic dam of Cublize (France).

Modeling processes from timed observations

This paper addresses the problem of learning a Dynamic Bayesian Network from timed data without prior knowledge to the system. One of the main problems of learning a Dynamic Bayesian Network is building and orienting the edges of the network avoiding loops. The problem is more difficult when data are timed. This paper proposes a new algorithm to learn the structure of a Dynamic Bayesian Network and to orient the edges from the timed data contained in a given timed data base. This algorithm is based on an adequate representation of a set of sequences of timed data and uses an information based measure of the relations between two edges. This algorithm is a part of the Timed Observation Mining for Learning (TOM4L) process that is based on the Theory of the Timed Observations. The paper illustrates the algorithm with a theoretical example before presenting the results on an application on the Apache system of the Arcelor- Mittal Steel Group, a real world knowledge based system that diagnoses a galvanization bath.

Learning dynamic bayesian networks with the tom4l process

Knowledge Engineering (KE) provides resources to build a conceptual model from experts’ knowledge which is sometimes deficient to interpret the input data flow coming from a concrete process On the other hand, data mining techniques in a process of Knowledge Discovery in Databases (KDD) can be used in order to obtain representative patterns of data which could allow to improve the model to be constructed However, interpreting these patterns is difficult due to the gap which exists between the expert’s conceptual universe and that of the process instrumentation This chapter proposes then a global approach which combines KE with KDD in order to allow the construction of Knowledge Models for Knowledge Based Systems from expert knowledge and knowledge discovered in data This approach is grounded in the Theory of Timed Observations on which both a KE methodology and a KDD process are based, so that the resulting models can be compared

Integrating Knowledge Engineering with Knowledge Discovery in Database: TOM4D and TOM4L

A dam diagnosis method based on a multimodel approach is proposed, and structural, functional, behavioral, and perception models are defined. The modeling process treats a dam as a connecte...

Marc Le Goc

Papers

Levels of Abstraction for Behavior Modeling in the GerHome Project

Modeling processes from timed observations

Learning dynamic bayesian networks with the tom4l process

Integrating Knowledge Engineering with Knowledge Discovery in Database: TOM4D and TOM4L

Multimodel-Based Diagnosis of Hydraulic Dams