Home About us Contact
|
Home About us Contact | ||
|
|
||
Learning Techniques (learning + techniques)
Kinds of Learning Techniques Selected AbstractsMONITORING BEHAVIOR WITH AN ARRAY OF SENSORSCOMPUTATIONAL INTELLIGENCE, Issue 4 2007Dorothy N. Monekosso The objective is to detect activities taking place in a home and to create a model of behavior for the occupant. A behavior is a pattern in the sequence of activities. An array of sensors captures the status of appliances. Models for the occupant's activities are built from the captured data using supervised and unsupervised learning techniques. The models of behavior are built using the hidden Markov model (HMM) technique. Predictive models can be used in a number of ways: to enhance user experience, to maximize resource usage efficiency, for safety and security. This work focuses on supporting independent living and enhancing quality of life of older persons. The ultimate goal is for the system to distinguish between normal and anomalous behavior. In this paper, we present the results of comparing supervised and unsupervised classification techniques applied to the problem of modeling activity for the purpose of modeling behavior in a home. [source] Financial decision support using neural networks and support vector machinesEXPERT SYSTEMS, Issue 4 2008Chih-Fong Tsai Abstract: Bankruptcy prediction and credit scoring are the two important problems facing financial decision support. The multilayer perceptron (MLP) network has shown its applicability to these problems and its performance is usually superior to those of other traditional statistical models. Support vector machines (SVMs) are the core machine learning techniques and have been used to compare with MLP as the benchmark. However, the performance of SVMs is not fully understood in the literature because an insufficient number of data sets is considered and different kernel functions are used to train the SVMs. In this paper, four public data sets are used. In particular, three different sizes of training and testing data in each of the four data sets are considered (i.e. 3:7, 1:1 and 7:3) in order to examine and fully understand the performance of SVMs. For SVM model construction, the linear, radial basis function and polynomial kernel functions are used to construct the SVMs. Using MLP as the benchmark, the SVM classifier only performs better in one of the four data sets. On the other hand, the prediction results of the MLP and SVM classifiers are not significantly different for the three different sizes of training and testing data. [source] Chaotic analysis of predictability versus knowledge discovery techniques: case study of the Polish stock marketEXPERT SYSTEMS, Issue 5 2002Hak Chun Increasing evidence over the past decade indicates that financial markets exhibit nonlinear dynamics in the form of chaotic behavior. Traditionally, the prediction of stock markets has relied on statistical methods including multivariate statistical methods, autoregressive integrated moving average models and autoregressive conditional heteroskedasticity models. In recent years, neural networks and other knowledge techniques have been applied extensively to the task of predicting financial variables. This paper examines the relationship between chaotic models and learning techniques. In particular, chaotic analysis indicates the upper limits of predictability for a time series. The learning techniques involve neural networks and case,based reasoning. The chaotic models take the form of R/S analysis to measure persistence in a time series, the correlation dimension to encapsulate system complexity, and Lyapunov exponents to indicate predictive horizons. The concepts are illustrated in the context of a major emerging market, namely the Polish stock market. [source] Concurrent Q-learning: Reinforcement learning for dynamic goals and environmentsINTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS, Issue 10 2005Robert B. Ollington This article presents a powerful new algorithm for reinforcement learning in problems where the goals and also the environment may change. The algorithm is completely goal independent, allowing the mechanics of the environment to be learned independently of the task that is being undertaken. Conventional reinforcement learning techniques, such as Q-learning, are goal dependent. When the goal or reward conditions change, previous learning interferes with the new task that is being learned, resulting in very poor performance. Previously, the Concurrent Q-Learning algorithm was developed, based on Watkin's Q-learning, which learns the relative proximity of all states simultaneously. This learning is completely independent of the reward experienced at those states and, through a simple action selection strategy, may be applied to any given reward structure. Here it is shown that the extra information obtained may be used to replace the eligibility traces of Watkin's Q-learning, allowing many more value updates to be made at each time step. The new algorithm is compared to the previous version and also to DG-learning in tasks involving changing goals and environments. The new algorithm is shown to perform significantly better than these alternatives, especially in situations involving novel obstructions. The algorithm adapts quickly and intelligently to changes in both the environment and reward structure, and does not suffer interference from training undertaken prior to those changes. © 2005 Wiley Periodicals, Inc. Int J Int Syst 20: 1037,1052, 2005. [source] Using Role-Play Scenarios in the IR Classroom: An Examination of Exercises on Peacekeeping Operations and Foreign Policy Decision MakingINTERNATIONAL STUDIES PERSPECTIVES, Issue 1 2004Carolyn M. Shaw Use of role-play scenarios in the classroom is just one of a number of active learning techniques that are being used more and more frequently to convey the more abstract concepts of international relations (IR) to students in a meaningful way. This paper examines the value of two specific role-play exercises used in an introduction to international relations course on the topics of peacekeeping and foreign policy decision making. The value of such interactive exercises is laid out in a section examining what learning objectives can be achieved by using role-play scenarios. These include promoting student interaction and input, and promoting student curiosity and creativity. The preparations necessary for conducting such an exercise are laid out, followed by a description of the exercises as they were conducted in the classroom. Finally, an assessment of the exercises provides useful feedback on the degree to which specific learning objectives were achieved, and how such exercises can be modified to be even more effective. [source] Toward a Global Theory of Mind: The Potential Benefits of Presenting a Range of IR Theories through Active LearningINTERNATIONAL STUDIES PERSPECTIVES, Issue 4 2003A. L. Morgan Active learning is particularly well-suited to teaching across the range of perspectives inherent in the practice and study of international politics for two key reasons: (1) because of its capacity to highlight how subjective, intersubjective, and contested understandings play an important role in determining outcomes in the ivory tower as well as in the real world and (2) because of the compatibility between underlying theories of knowledge that inform active learning and the newer generation of IR theories including subaltern realism, social constructivism, constitutive theory, and postmodernism. This article explores the potential benefits of presenting these and other norm-oriented theories through active learning. It also discusses ways to overcome barriers to the integration of active learning techniques. [source] Continuing Medical Education, Continuing Professional Development, and Knowledge Translation: Improving Care of Older Patients by Practicing PhysiciansJOURNAL OF AMERICAN GERIATRICS SOCIETY, Issue 10 2006David C. Thomas MD Many community-based internists and family physicians lack familiarity with geriatrics knowledge and best practices, but they face overwhelming fiscal and time barriers to expanding their skills and improving their behavior in the care of older people. Traditional lecture-and-slide-show continuing medical education (CME) programs have been shown to be relatively ineffective in changing this target group's practice. The challenge for geriatrics educators, then, is to devise CME programs that are highly accessible to practicing physicians, that will have an immediate and significant effect on practitioners' behavior, and that are financially viable. Studies of CME have shown that the most effective programs for knowledge translation in these circumstances involve what is known as active-mode learning, which relies on interactive, targeted, and multifaceted techniques. A systematic literature review, supplemented by structured interviews, was performed to inventory active-mode learning techniques for geriatrics knowledge and skills in the United States. Thirteen published articles met the criteria, and leaders of 28 active-mode CME programs were interviewed. This systematic review indicates that there is a substantial experience in geriatrics training for community-based physicians, much of which is unpublished and incompletely evaluated. It appears that the most effective methods to change behaviors involved multiple educational efforts such as written materials or toolkits combined with feedback and strong communication channels between instructors and learners. [source] Early elementary students' development of astronomy concepts in the planetariumJOURNAL OF RESEARCH IN SCIENCE TEACHING, Issue 2 2009Julia D. Plummer Abstract The National Science Education Standards [National Research Council (1996) National science education standards. Washington, DC: National Academy Press] recommend that students understand the apparent patterns of motion of the sun, moon and stars by the end of early elementary school. However, little information exists on students' ability to learn these concepts. This study examines the change in students' understanding of apparent celestial motion after attending a planetarium program using kinesthetic learning techniques. Pre- and post-interviews were conducted with participants from seven classes of first and second grade students (N,=,63). Students showed significant improvement in knowledge of all areas of apparent celestial motion covered by the planetarium program. This suggests that students in early elementary school are capable of learning the accurate description of apparent celestial motion. The results also demonstrate the value of both kinesthetic learning techniques and the rich visual environment of the planetarium for improved understanding of celestial motion. © 2008 Wiley Periodicals, Inc. J Res Sci Teach 46: 192,209, 2009 [source] Automating survey coding by multiclass text categorization techniquesJOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE AND TECHNOLOGY, Issue 14 2003Daniela Giorgetti Survey coding is the task of assigning a symbolic code from a predefined set of such codes to the answer given in response to an open-ended question in a questionnaire (aka survey). This task is usually carried out to group respondents according to a predefined scheme based on their answers. Survey coding has several applications, especially in the social sciences, ranging from the simple classification of respondents to the extraction of statistics on political opinions, health and lifestyle habits, customer satisfaction, brand fidelity, and patient satisfaction. Survey coding is a difficult task, because the code that should be attributed to a respondent based on the answer she has given is a matter of subjective judgment, and thus requires expertise. It is thus unsurprising that this task has traditionally been performed manually, by trained coders. Some attempts have been made at automating this task, most of them based on detecting the similarity between the answer and textual descriptions of the meanings of the candidate codes. We take a radically new stand, and formulate the problem of automated survey coding as a text categorization problem, that is, as the problem of learning, by means of supervised machine learning techniques, a model of the association between answers and codes from a training set of precoded answers, and applying the resulting model to the classification of new answers. In this article we experiment with two different learning techniques: one based on naive Bayesian classification, and the other one based on multiclass support vector machines, and test the resulting framework on a corpus of social surveys. The results we have obtained significantly outperform the results achieved by previous automated survey coding approaches. [source] FLOOD STAGE FORECASTING WITH SUPPORT VECTOR MACHINES,JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION, Issue 1 2002Shie-Yui Liong ABSTRACT: Machine learning techniques are finding more and more applications in the field of forecasting. A novel regression technique, called Support Vector Machine (SVM), based on the statistical learning theory is explored in this study. SVM is based on the principle of Structural Risk Minimization as opposed to the principle of Empirical Risk Minimization espoused by conventional regression techniques. The flood data at Dhaka, Bangladesh, are used in this study to demonstrate the forecasting capabilities of SVM. The result is compared with that of Artificial Neural Network (ANN) based model for one-lead day to seven-lead day forecasting. The improvements in maximum predicted water level errors by SVM over ANN for four-lead day to seven-lead day are 9.6 cm, 22.6 cm, 4.9 cm and 15.7 cm, respectively. The result shows that the prediction accuracy of SVM is at least as good as and in some cases (particularly at higher lead days) actually better than that of ANN, yet it offers advantages over many of the limitations of ANN, for example in arriving at ANN's optimal network architecture and choosing useful training set. Thus, SVM appears to be a very promising prediction tool. [source] Appraising and assessing reflection in students' writing on a structured worksheetMEDICAL EDUCATION, Issue 6 2002Barbel Pee Background A variety of teaching and learning techniques intended to engage students in reflection are either in use or are being developed in medical and dental education. In line with evidence-based practice in education, research is needed to appraise the utility and effectiveness of these techniques, so that they may be used with confidence. Aim To assess whether students completing a `reflective' learning activity based on a structured worksheet really were reflecting. Method, A qualitative, multi-method approach was taken. Worksheets completed by students were examined for evidence of reflection by researchers using two sets of criteria for the assessment of reflection derived from the literature, and by peer judges using their own criteria. The opinions of students completing the activity, regarding its acceptability and utility, were elicited by a questionnaire incorporating a 5-point Likert scale. Results Results from all methods suggest that students completing the activity were reflecting. Students' opinions of the activity were mainly positive. Conclusion, The methods employed may be of use to educators wishing to appraise reflective learning activities or, possibly, to assess student reflection. [source] Predictions of peptides' retention times in reversed-phase liquid chromatography as a new supportive tool to improve protein identification in proteomicsPROTEINS: STRUCTURE, FUNCTION AND BIOINFORMATICS, Issue 4 2009Tomasz B, czek Dr. Abstract One of the initial steps of proteomic analysis is peptide separation. However, little information from RP-HPLC, employed for peptides separation, is utilized in proteomics. Meanwhile, prediction of the retention time for a given peptide, combined with routine MS/MS data analysis, could help to improve the confidence of peptide identifications. Recently, a number of models has been proposed to characterize quantitatively the structure of a peptide and to predict its gradient RP-HPLC retention at given separation conditions. The chromatographic behavior of peptides has usually been related to their amino acid composition. However, different values of retention coefficients of the same amino acid in different peptides at different neighborhoods were commonly observed. Therefore, specific retention coefficients were derived by regression analysis or by artificial neural networks (ANNs) with the use of a set of peptides retention. In the review, various approaches for peptide elution time prediction in RP-HPLC are presented and critically discussed. The contribution of sequence dependent parameters (e.g., amphipathicity or peptide sequence) and peptide physicochemical descriptors (e.g., hydrophobicity or peptide length) that have been shown to affect the peptide retention time in LC are considered and analyzed. The predictive capability of the retention time prediction models based on quantitative structure,retention relationships (QSRRs) are discussed in details. Advantages and limitations of various retention prediction strategies are identified. It is concluded that proper processing of chromatographic data by statistical learning techniques can result in information of direct use for proteomics, which is otherwise wasted. [source] Proteomic patterns for classification of ovarian cancer and CTCL serum samples utilizing peak pairs indicative of post-translational modificationsPROTEINS: STRUCTURE, FUNCTION AND BIOINFORMATICS, Issue 22 2007Chenwei Liu Abstract Proteomic patterns as a potential diagnostic technology has been well established for several cancer conditions and other diseases. The use of machine learning techniques such as decision trees, neural networks, genetic algorithms, and other methods has been the basis for pattern determination. Cancer is known to involve signaling pathways that are regulated through PTM of proteins. These modifications are also detectable with high confidence using high-resolution MS. We generated data using a prOTOFÔ mass spectrometer on two sets of patient samples: ovarian cancer and cutaneous t-cell lymphoma (CTCL) with matched normal samples for each disease. Using the knowledge of mass shifts caused by common modifications, we built models using peak pairs and compared this to a conventional technique using individual peaks. The results for each disease showed that a small number of peak pairs gave classification equal to or better than the conventional technique that used multiple individual peaks. This simple peak picking technique could be used to guide identification of important peak pairs involved in the disease process. [source] Teaching Across the Generation Gap: A Consensus from the Council of Emergency Medicine Residency Directors 2009 Academic AssemblyACADEMIC EMERGENCY MEDICINE, Issue 2009Lisa Moreno-Walton MD Abstract Background:, Four distinct generations of physicians currently coexist within the emergency medicine (EM) workforce, each with its own unique life experience, perspective, attitude, and expectation of work and education. To the best of our knowledge, no investigations or consensus statements exist that specifically address the effect of intergenerational differences on undergraduate and graduate medical education in EM. Objectives:, To review the existing literature on generational differences as they pertain to workforce expectations, educational philosophy, and learning styles and to create a consensus statement based on the shared insights of experienced educators in EM, with specific recommendations to improve the effectiveness of EM residency training programs. Methods:, A group of approximately one hundred EM program directors (PDs), assistant PDs, and other academic faculty attending an annual conference of emergency physician (EP) educators gathered at a breakout session and working group to examine the literature on intergenerational differences, to share insights and discuss interventions tailored to address these stylistic differences, and to formulate consensus recommendations. Results:, A set of specific recommendations, including effective educational techniques, was created based on literature from other professions and medical disciplines, as well as the contributions of a diverse group of EP educators. Conclusions:, Recommendations included early establishment of clear expectations and consequences, emphasis on timely feedback and individualized guidance during training, explicit reinforcement of a patient-centered care model, use of peer modeling and support, and emphasis on more interactive and small-group learning techniques. [source] Problem-Based Learning Biotechnology Courses in Chemical EngineeringBIOTECHNOLOGY PROGRESS, Issue 1 2006Charles E. Glatz We have developed a series of upper undergraduate/graduate lecture and laboratory courses on biotechnological topics to supplement existing biochemical engineering, bioseparations, and biomedical engineering lecture courses. The laboratory courses are based on problem-based learning techniques, featuring two- and three-person teams, journaling, and performance rubrics for guidance and assessment. Participants initially have found them to be difficult, since they had little experience with problem-based learning. To increase enrollment, we are combining the laboratory courses into 2-credit groupings and allowing students to substitute one of them for the second of our 2-credit chemical engineering unit operations laboratory courses. [source] | ||||||||||||||||