Results 11  20
of
281
General and Efficient Multisplitting of Numerical Attributes
, 1999
"... Often in supervised learning numerical attributes require special treatment and do not fit the learning scheme as well as one could hope. Nevertheless, they are common in practical tasks and, therefore, need to be taken into account. We characterize the wellbehavedness of an evaluation function, a ..."
Abstract

Cited by 53 (7 self)
 Add to MetaCart
(Show Context)
Often in supervised learning numerical attributes require special treatment and do not fit the learning scheme as well as one could hope. Nevertheless, they are common in practical tasks and, therefore, need to be taken into account. We characterize the wellbehavedness of an evaluation function, a property that guarantees the optimal multipartition of an arbitrary numerical domain to be defined on boundary points. Wellbehavedness reduces the number of candidate cut points that need to be examined in multisplitting numerical attributes. Many commonly used attribute evaluation functions possess this property; we demonstrate that the cumulative functions Information Gain and Training Set Error as well as the noncumulative functions Gain Ratio and Normalized Distance Measure are all wellbehaved. We also devise a method of finding optimal multisplits efficiently by examining the minimum number of boundary point combinations that is required to produce partitions which are optimal wit...
Simplifying Decision Trees: A Survey
, 1996
"... Induced decision trees are an extensivelyresearched solution to classification tasks. For many practical tasks, the trees produced by treegeneration algorithms are not comprehensible to users due to their size and complexity. Although many tree induction algorithms have been shown to produce simpl ..."
Abstract

Cited by 47 (6 self)
 Add to MetaCart
Induced decision trees are an extensivelyresearched solution to classification tasks. For many practical tasks, the trees produced by treegeneration algorithms are not comprehensible to users due to their size and complexity. Although many tree induction algorithms have been shown to produce simpler, more comprehensible trees (or data structures derived from trees) with good classification accuracy, tree simplification has usually been of secondary concern relative to accuracy and no attempt has been made to survey the literature from the perspective of simplification. We present a framework that organizes the approaches to tree simplification and summarize and critique the approaches within this framework. The purpose of this survey is to provide researchers and practitioners with a concise overview of treesimplification approaches and insight into their relative capabilities. In our final discussion, we briefly describe some empirical findings and discuss the application of tree i...
Adaptive fastest path computation on a road network: A traffic mining approach
 In Proc. 2007 Int. Conf. on Very Large Data Bases (VLDB’07
, 2007
"... Efficient fastest path computation in the presence of varying speed conditions on a large scale road network is an essential problem in modern navigation systems. Factors affecting road speed, such as weather, time of day, and vehicle type, need to be considered in order to select fast routes that m ..."
Abstract

Cited by 46 (2 self)
 Add to MetaCart
(Show Context)
Efficient fastest path computation in the presence of varying speed conditions on a large scale road network is an essential problem in modern navigation systems. Factors affecting road speed, such as weather, time of day, and vehicle type, need to be considered in order to select fast routes that match current driving conditions. Most existing systems compute fastest paths based on road Euclidean distance and a small set of predefined road speeds. However, “History is often the best teacher”. Historical traffic data or driving patterns are often more useful than the simple Euclidean distancebased computation because people must have good reasons to choose these routes, e.g., they may want to avoid those that pass through high crime areas at night or that likely encounter accidents, road construction, or traffic jams. In this paper, we present an adaptive fastest path algorithm capable of efficiently accounting for important driving and speed patterns mined from a large set of traffic data. The algorithm is based on the following observations: (1) The hierarchy of roads can be used to partition the road network into areas, and different path precomputation strategies can be used at the area level, (2) we can limit our route search strategy to edges and path segments that are actually frequently traveled in the data, and (3) drivers usually traverse the road network through the largest roads available given the distance of the trip, except if there are small roads with a significant speed advantage over the large ones. Through an extensive experimental evaluation on real road networks we show that our algorithm provides desirable (short and wellsupported) routes, and that it is significantly faster than competing methods.
Efficient C4.5
, 2000
"... We present an analytic evaluation of the runtime behavior of the C4.5 algorithm which highlights some efficiency improvements. We have implemented a more efficient version of the algorithm, called EC4.5, that improves on C4.5 by adopting the best among three strategies at each node construction. ..."
Abstract

Cited by 38 (4 self)
 Add to MetaCart
We present an analytic evaluation of the runtime behavior of the C4.5 algorithm which highlights some efficiency improvements. We have implemented a more efficient version of the algorithm, called EC4.5, that improves on C4.5 by adopting the best among three strategies at each node construction. The first strategy uses a binary search of thresholds instead of the linear search of C4.5. The second strategy adopts a counting sort method instead of the quicksort of C4.5. The third strategy uses a mainmemory version of the RainForest algorithm for constructing decision trees. Our implementation computes the same decision trees as C4.5 with a performance gain of up to 5 times.
MRDTL: A multirelational decision tree learning algorithm
 Proceedings of the 13th International Conference on Inductive Logic Programming (ILP 2003
, 2002
"... this paper, we have described an implementation of multirelational decision tree learning (MRDTL) algorithm based on the techniques proposed by Knobbe et el. (Knobbe et el., 1999a, Knobbe et el., 1999b) ..."
Abstract

Cited by 37 (2 self)
 Add to MetaCart
(Show Context)
this paper, we have described an implementation of multirelational decision tree learning (MRDTL) algorithm based on the techniques proposed by Knobbe et el. (Knobbe et el., 1999a, Knobbe et el., 1999b)
Volume Under the ROC Surface for Multiclass Problems. Exact Computation and Evaluation of Approximations
 Proc. of 14th European Conference on Machine Learning
, 2003
"... Receiver Operating Characteristic (ROC) has been successfully applied to classifier problems with two classes. The Area Under the ROC Curve (AUC) has been determined as a better way to evaluate classifiers than predictive accuracy or error. However, the extension of the Area Under the ROC Curve f ..."
Abstract

Cited by 35 (0 self)
 Add to MetaCart
(Show Context)
Receiver Operating Characteristic (ROC) has been successfully applied to classifier problems with two classes. The Area Under the ROC Curve (AUC) has been determined as a better way to evaluate classifiers than predictive accuracy or error. However, the extension of the Area Under the ROC Curve for more than two classes has not been addressed to date, because of the complexity and elusiveness of its precise definition. In this paper, we present the real extension to the Area Under the ROC Curve in the form of the Volume Under the ROC Surface (VUS), showing how to compute the polytope that corresponds to the absence of classifiers (given only by the trivial classifiers), to the best classifier and to whatever set of classifiers. We compare the real VUS with "approximations"or "extensions"of the AUC for more than two classes.
Analysis of interpretabilityaccuracy tradeoff of fuzzy systems by multiobjective fuzzy geneticsbased machine learning
 International Journal of Approximate Reasoning
, 2007
"... This paper examines the interpretabilityaccuracy tradeoff in fuzzy rulebased classifiers using a multiobjective fuzzy geneticsbased machine learning (GBML) algorithm. Our GBML algorithm is a hybrid version of Michigan and Pittsburgh approaches, which is implemented in the framework of evolutionar ..."
Abstract

Cited by 35 (7 self)
 Add to MetaCart
(Show Context)
This paper examines the interpretabilityaccuracy tradeoff in fuzzy rulebased classifiers using a multiobjective fuzzy geneticsbased machine learning (GBML) algorithm. Our GBML algorithm is a hybrid version of Michigan and Pittsburgh approaches, which is implemented in the framework of evolutionary multiobjective optimization (EMO). Each fuzzy rule is represented by its antecedent fuzzy sets as an integer string of fixed length. Each fuzzy rulebased classifier, which is a set of fuzzy rules, is represented as a concatenated integer string of variable length. Our GBML algorithm simultaneously maximizes the accuracy of rule sets and minimizes their complexity. The accuracy is measured by the number of correctly classified training patterns while the complexity is measured by the number of fuzzy rules and/or the total number of antecedent conditions of fuzzy rules. We examine the interpretabilityaccuracy tradeoff for training patterns through computational experiments on some benchmark data sets. A clear tradeoff structure is visualized for each data set. We also examine the interpretabilityaccuracy tradeoff for test patterns. Due to the overfitting to training patterns, a clear tradeoff structure is not always obtained in computational experiments for test patterns.
Diagnosing and acting on student affect: the tutor's perspective.” User Modeling and UserAdapted Interaction
, 2008
"... Abstract. In this paper we explore human tutors ’ inferences in relation to learners ’ affective states and the relationship between those inferences and the actions that tutors take as their consequence. At the core of our investigations presented in this paper lie fundamental questions associated ..."
Abstract

Cited by 30 (6 self)
 Add to MetaCart
Abstract. In this paper we explore human tutors ’ inferences in relation to learners ’ affective states and the relationship between those inferences and the actions that tutors take as their consequence. At the core of our investigations presented in this paper lie fundamental questions associated with the notion of affect: its nature, its role in education in general, and in particular its impact on the way successful humancomputer communication and education can be facilitated. A theoretical framework is presented which serves as the basis for determining the factors that may be relevant to tutors ’ decisions. A study was designed to determine what affective states of the student are relevant to tutoring mathematics and for building a predictive model of affect from tutor perspective. Logs of tutorstudent dialogue were annotated in order to determine the types and range of student and tutor actions. Machine learning techniques were then applied to those actions to predict the values of the factors most relevant to tutor actions. Specifically, The results for the analysis of three factors related to student affect: student confidence, student interest and student effort suggest that machine learning techniques were useful in identifying the possible correspondences between the three factors and tutor actions, but not sufficiently reliable when based solely on the data collected, due to the limited size and sparsity of the dataset. However, the results are considered as the basis for empirically derived hypotheses to be tested in further studies. The potential implications of the hypotheses, if they were confirmed by further studies, are discussed in relation to understanding and supporting humancomputer interactions in educational settings.
Zeta: A Global Method for Discretization of Continuous Variables
 Proc. Third Int’l Conf. Knowledge Discovery and Data Mining (KDD97
, 1997
"... Discretization of continuous variables so they may be used in conjunction with machine learning or statistical techniques that require nominal data is an important problem to be solved in developing generally applicable methods for data mining. This paper introduces a new technique for discretizatio ..."
Abstract

Cited by 30 (3 self)
 Add to MetaCart
(Show Context)
Discretization of continuous variables so they may be used in conjunction with machine learning or statistical techniques that require nominal data is an important problem to be solved in developing generally applicable methods for data mining. This paper introduces a new technique for discretization of such variables based on zeta, a measure of strength of association between nominal variables developed for this purpose. Following a review of existing techniques for discretization we define zeta, a measure based on minimisation of the error rate when each value of an independent variable must predict a different value of a dependent variable. We then describe both how a continuous variable may be dichotomised by searching for a maximum value of zeta, and how a heuristic extension of this method can partition a continuous variable into more than two categories. A series of experimental evaluations of zetadiscretization, including comparisons with other published methods, show that zetadiscretization runs considerably faster than other techniques without any loss of accuracy. We conclude that zetadiscretization offers considerable advantages over alternative procedures and discuss some of the ways in which it could be enhanced. 1
Decision Forest: Combining the Predictions of Multiple Independent Decision Tree Models
 J Chem Inf Comput Sci
"... The techniques of combining the results of multiple classification models to produce a single prediction have been investigated for many years. In earlier applications, the multiple models to be combined were developed by altering the training set. The use of these socalled resampling techniques, h ..."
Abstract

Cited by 27 (10 self)
 Add to MetaCart
The techniques of combining the results of multiple classification models to produce a single prediction have been investigated for many years. In earlier applications, the multiple models to be combined were developed by altering the training set. The use of these socalled resampling techniques, however, poses the risk of reducing predictivity of the individual models to be combined and/or over fitting the noise in the data, which might result in poorer prediction of the composite model than the individual models. In this paper, we suggest a novel approach, named Decision Forest, that combines multiple Decision Tree models. Each Decision Tree model is developed using a unique set of descriptors. When models of similar predictive quality are combined using the Decision Forest method, quality compared to the individual models is consistently and significantly improved in both training and testing steps. An example will be presented for prediction of binding affinity of 232 chemicals to the estrogen receptor.