An algorithm is considered which for a given decision table constructs a decision tree with minimal depth. The class of all information systems (finite and infinite) is described for which this algorithm has polynomial time complexity depending on the number of columns (attributes) in decision tables.
The analysis of the research problem started from listing the issues of decision theory and the decision-making process that support the process of building decision trees. The area in question covers a procedure for building and proceeding when creating decision trees. The solution to the research problem consists in defining decision variables and arranging them into logical statements by writing down all possible variants and only accounting for the true ones. True solutions derived from coding were detailed and the number of occurring decision trees was calculated in the case under consideration. The decision problem was presented in the form of decision trees, which made it possible to select the optimum decision tree. The obtained results were considered and the optimum decision tree was chosen. At the same time, the record of decision variables was analyzed, providing the answer as to which courier company will best meet expectations of entrepreneurs and ensure the most satisfying cooperation. That company turned out to be K-EX. The article aimed to select a courier company from the perspective of online retailers, with the selection having been made using the method of decision trees based on four basic criteria defined within the research.
Marketing campaigns directed to randomly selected customers often generate huge costs and a weak response. Moreover, such campaigns tend to unnecessarily annoy customers and make them less likely to answer to future communications. Precise targeting of marketing actions can potentially results in a greater return on investment. Usually, response models are used to select good targets. They aim at achieving high prediction accuracy for the probability of purchase based on a sample of customers, to whom a pilot campaign has been sent. However, to separate the impact of the action from other stimuli and spontaneous purchases we should model not the response probabilities themselves, but instead, the change in those probabilities caused by the action. The problem of predicting this change is known as uplift modeling, differential response analysis, or true lift modeling. In this work, tree-based classifiers designed for uplift modeling are applied to real marketing data and compared with traditional response models, and other uplift modeling techniques described in literature. The experiments show that the proposed approaches outperform existing uplift modeling algorithms and demonstrate significant advantages of uplift modeling over traditional, response based targeting.
4
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Drzewa decyzyjne są dogodną metodą strukturalizacji procesów decyzyjnych, w szczególności procesów sekwencyjnych podejmowanych w warunkach niepewności. Metoda drzew decyzyjnych pozwala także na ocenę efektywności i ryzyka badanych kierunków działań i wybór optymalnej decyzji. W artykule zaprezentowano przykład zastosowania metody w badaniu efektywności i ryzyka zagospodarowania złóż miedzi na dużych głębokościach.
EN
Decision trees constitute a convenient method for structuring decision-making processes, particulary sequential ones taken in the conditions of uncertainty. This method enables also evaluation of effectiveness and risk of the actions considered and the choice of an optimal decision. The paper presents an example of its application while studying effectiveness and risk associated with development of deep copper deposits.
Basing on the review of available literature devoted to applications of programming tools in research on mechanism of music perception, the first results of my own research devoted to mining of hidden regularities in pieces of various types of music is presented. General characteristics of the developed databases, methodology of executed research, and discussion of the discovered knowledge structures for selected types of music pieces are briefly dealt with.
6
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The cluster analysis is applied to the analysis of the data describing the status of protein structure in respect to hydrophobic core characteristics. The analysis revealed presence of two clusters distinguishing the proteins accordant with the “fuzzy oil drop” model and those which appear as discordant in respect to this model. The analysis was performed separately for chains treated as structural unit and for units defined according to IV-order (taking the functional protein complex). The characteristics of these two classification system appeared to differ in respect to number of proteins belonging to each of two clusters as well as relation between them.
7
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
In this paper the problem of recognition of words with fixed length from a regular language is considered. The word under consideration can be interpreted as a description of certain screen image in the following way: the i-th letter of the word encodes the color of the i-th screen cell. In this case a decision tree which recognizes some words may be interpreted as an algorithm for the recognition of images which are defined by considered words. The classification of all regular languages depending on the growth of minimal depth of decision trees for language word recognition with the growth of the word length is obtained. In proofs methods of test theory and rough set theory are used.
8
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The main task in decision tree construction algorithms is to find the "best partition" of the set of objects. In this paper, we investigate the problem of optimal binary partition of continous attribute domain for large data sets stored in relational databases. The critical for time complexity of algorithms solving this problem is the number of simple SQL queries necassary to construct such partitions. Using straightforward approach to optimal partition selection the number of necessary queries is of order O(N), where N is the number of preassumed partitions of the searching space. We show some properties of optimization measures related to discernibility between objects, that allow to reduce the size of searching space. We prove that using only O(log N) simple queries, one can construct the partition vey close to optimal.
Complex risky decision problems involve sequences of decisions and random events. The choice at a given stage depends on the decisions taken in the previous stages, as well as on the realizations of the random events that occurred earlier. In the analysis of such situations, decision trees are used, and the criterion for choosing the optimal decision is to maximize the expected monetary value. Unfortunately, this approach often does not reflect the actual choices of individual decision makers. In descriptive decision theory, the criterion of maximizing the expected monetary value is replaced by a subjective valuation that takes into account the relative outcomes and their probabilities. This paper presents a proposal to use the principles of cumulative prospect theory to analyse complex decision problems. The concept of a certainty equivalent is used to make it possible to compare risky and non-risky alternatives.
In this paper a single-objective Genetic Algorithm is exploited to optimise a Fuzzy Decision Tree for fault classification. The optimisation procedure is presented with respect to an ancillary classification problem built with artificial data. Work is in progress for the application of the proposed approach to a real fault classification problem.
Complex risky decision problems involve sequences of decisions and random events. The choice at a given stage depends on the decisions taken in the previous stages, as well as on the realizations of the random events that occurred earlier. In the analysis of such situations, decision trees are used, and the criterion for choosing the optimal decision is to maximize the expected monetary value. Unfortunately, this approach often does not reflect the actual choices of individual decision makers. In descriptive decision theory, the criterion of maximizing the expected monetary value is replaced by a subjective valuation that takes into account the relative outcomes and their probabilities. This paper presents a proposal to use the principles of cumulative prospect theory to analyse complex decision problems. The concept of a certainty equivalent is used to make it possible to compare risky and non-risky alternatives.
12
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
An approximate algorithm for minimization of weighted depth of decision trees is considered. A bound on accuracy of this algorithm is obtained which is unimprovable in general case. Under some natural assumptions on the class NP, the considered algorithm is close (from the point of view of accuracy) to best polynomial approximate algorithms for minimization of weighted depth of decision trees.
13
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Active learning is the process in which unlabeled instances are dynamically selected for expert labelling, and then a classifier is trained on the labeled data. Active learning is particularly useful when there is a large set of unlabeled instances, and acquiring a label is costly. In business scenarios such as direct marketing, active learning can be used to indicate which customer to approach such that the potential benefit from the approached customer can cover the cost of approach. This paper presents a new algorithm for cost-sensitive active learning using a conditional expectation estimator. The new estimator focuses on acquisitions that are likely to improve the profit. Moreover, we investigate simulated annealing techniques to combine exploration with exploitation in the classifier construction. Using five evaluation metrics, we evaluated the algorithm on four benchmark datasets. The results demonstrate the superiority of the proposed method compared to other algorithms.
14
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
This paper presents a new tool for the study of relationships between the total path length or the average depth and the number of misclassifications for decision trees. In addition to algorithm, the paper also presents the results of experiments with datasets from UCI ML Repository [9] and datasets representing Boolean functions with 10 variables.
16
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Decision trees arę one of the computing intelligence methods which proved to be very reliable as far as solving complicated multidimensional problems is concerned. Therefore, these methods are often used for extracting rules and to predict variables, what makes them useful for production automation. In this paper authors discuss the possibility of the use of decision trees for electric are steelmaking process. The main goal is to predict temperature in the electric are furnace by the use of decision trees. Proper automatic temperature prediction may reduce the number of temperature measurements during the process and consequently, it may shorten the time of the process. Optimization of production processes leads to real benefits, which is, for example, lowering costs of production. Calculations were done by the use of six types of regression decisions trees available in Statistica Data Miner software. The algorithms were examined considering the minimum error rate of temperature prediction, but also less complicated tree structure. The structure of a decision tree is also important owing to computational complexity.
PL
Drzewa decyzji są jednymi z metod inteligencji obliczeniowej, które okazały się niezawodne przy rozwiązywaniu skomplikowanych, wielowymiarowych problemów obliczeniowych. Dlatego też, metody te są często stosowane do ekstrakcji reguł oraz do przewidywania wartości zmiennych, co czyni je szczególnie użytecznymi w problemach automatyzacji produkcji. W niniejszej pracy autorzy zaprezentują możliwość zastosowania drzew decyzji podczas procesu elektrołukowego. Głównym celem jest predykcja temperatury w elektrycznym piecu łukowym przy użyciu drzew decyzji. Poprawne i automatyczne przewidywanie temperatury może pozwolić na redukcję liczby wykonywanych pomiarów podczas procesu, a co za tym idzie, może pozwolić na skrócenie czasu całego procesu. Optymalizacja procesu daje wymierne korzyści, którymi mogą być na przykład niższe koszty produkcji. Obliczenia wykonane zostały przy użyciu sześciu typów drzew regresyjnych dostępnych w pakiecie Statistica Data Miner. Algorytmy były testowane pod względem osiągania jak najmniejszego błędu predykcji temperatury, ale także pod względem jak najmniej skomplikowanej struktury drzewa, która jest także ważnym elementem pod względem złożoności obliczeniowej.
Artificial intelligence is a field that has been rapidly developing in various areas of knowledge in recent years. Its application in medicine can support the intensive development of research in health care and improve and ac-celerate the operation of many medical facilities. This article presents sev-eral examples of expert systems that can find application in diagnosing and preparing a patient for selected tests. Expert systems can also find appli-cation in the rapid selection of rehabilitation, medical or support equip-ment and devices with which medical facilities are supplied. In this article, the reader will also find a sample application that will perform this func-tion. The article presents the elements of which a correct expert system should consist. For each application, tests have been carried out to show the correctness of the system. The purpose of the article was to show the capabilities of the expert system and its application in medical fields.
18
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
In the paper, we study a greedy algorithm for construction of decision trees. This algorithm is applicable to decision tables with many-valued decisions where each row is labeled with a set of decisions. For a given row, we should find a decision from the set attached to this row. Experimental results for data sets from UCI Machine Learning Repository and randomly generated tables are presented. We make a comparative study of the depth and average depth of the constructed decision trees for proposed approach and approach based on generalized decision. The obtained results show that the proposed approach can be useful from the point of view of knowledge representation and algorithm construction.
Hardware implementation of a widely used decision tree classifier is presented in this paper. The classifier task is to perform image-based object classification. The performance evaluation of the implemented architecture in terms of resource utilization and processing speed are reported. The presented architecture is compact, flexible and highly scalable and compares favorably to software-only solutions in terms of processing speed and power consumption.
This paper presents a computer-based method for recognizing digital images of bacterial cells. It covers automatic recognition of twenty genera and species of bacteria chosen by the author whose original contribution to the work consisted in the decision to conduct the process of recognizing bacteria using the simultaneous analysis of the following physical features of bacterial cells: color, size, shape, number of clusters, cluster shape, as well as density and distribution of the cells. The proposed method may be also used to recognize the microorganisms other than bacteria. In addition, it does not require the use of any specialized equipment. The lack of demand for high infrastructural standards and complementarity with the hardware and software widens the scope of the method’s application in diagnostics, including microbiological diagnostics. The proposed method may be used to identify new genera and species of bacteria, but also other microorganisms that exhibit similar morphological characteristics.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.