The classical rough set theory is based on the conventional indiscernibility relation. It is not very good for analyzing incomplete information. Some successful extended rough set models based on different non-equivalence relations have been proposed. The valued tolerance relation is such an extended model of classical rough set theory. However, the general calculation method of tolerance degree needs to know the prior probability distribution of an information system in advance, and it is also difficult to select a suitable threshold. In this paper, a data-driven valued tolerance relation (DVT) is proposed to solve this problem based on the idea of data-driven data mining. The new calculation method of tolerance degree and the auto-selection method of threshold do not require any prior domain knowledge except the data set. Some properties about the DVT are analyzed. Experiment results show that the DVT can get better and more stable classification results than other extended models of the classical rough set theory.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Recent developments in computing, communications, digital storage technologies, and high-throughput data-acquisition technologies, make it possible to gather and store incredible volumes of data. It creates unprecedented opportunities for knowledge discovery large-scale database. Data mining technology is a useful tool for this task. It is an emerging area of computational intelligence that offers new theories, techniques, and tools for processing large volumes of data, such as data analysis, decision making, etc. There are countless researchers working on designing efficient data mining techniques, methods, and algorithms. Unfortunately,most data mining researchers pay much attention to technique problems for developing data mining models and methods, while little to basic issues of data mining. What is data mining? What is the product of a data mining process? What are we doing in a data mining process? What is the rule we would obey in a data mining process? What is the relationship between the prior knowledge of domain experts and the knowledgemind from data? In this paper, we will address these basic issues of data mining from the viewpoint of informatics[1]. Data is taken as a manmade format for encoding knowledge about the natural world. We take data mining as a process of knowledge transformation. A domain-oriented data-driven data mining (3DM) model based on a conceptual data mining model is proposed. Some data-driven data mining algorithms are also proposed to show the validity of this model, e.g., the data-driven default rule generation algorithm, data-driven decision tree pre-pruning algorithm and data-driven knowledge acquisition from concept lattice.
The data-driven approach is a strong trend in modern marketing and other areas of life. From the position of methodological criticism, I discuss the data-driven mind traps. I indicate that contrary to popular beliefs: data does not speak for themselves; data is not given, but collected and generated, so biased; a larger amount of data does not necessarily mean a higher value; the tools used shape the examined reality; faith in data is a belief that is impossible to justify logically; there is no imperative but trend for the use of data and the most modern quantitative methods. I point to the links between epistemological and ethical assumptions as well as social consequences. I suggest alternatives
PL
Podejście data-driven jest silnym trendem we współczesnym marketingu i innych dziedzinach życia. Z pozycji krytyki metodologicznej autor omawia pułapki myślowe data-driven. Wskazuje, że wbrew pozorom, dane nie mówią same za siebie, dane nie są „dane”, tylko zbierane i generowane, więc nieobiektywne. Większa ilość danych nie musi oznaczać większej wartości. Stosowane narzędzia kształtują badaną rzeczywistość. Wiara w dane jest wiarą, niemożliwą do zasadniczo logicznego uzasadnienia. Panuje nie konieczność, ale moda na korzystanie z danych i najnowocześniejszych metod obliczeniowych. Autor wskazuje powiązania założeń epistemologicznych z etycznymi oraz konsekwencjami społecznymi oraz proponuje alternatywy.
4
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Purpose: The aims of this paper are to identify the root cause of the temporary decrease in the damping force which occurs during the early stage of the stroking cycle’s compression phase, the so-called damping lag, to describe measures of the phenomenon and to present methods for optimizing the design towards minimizing this (negative) effect. Design/methodology/approach: A theoretical background is presented in a constructive and computable manner with emphasis on data-driven modeling. The Design For Six Sigma (DFFS) approach and tools were used to validate the model statistically and, more importantly, to propose a method for data-driven optimization of the design. Findings: The root cause of the damping lag was confirmed during model validation as being a result of oil aeration. DFFS methodology proved to be useful in achieving design optimality. Research limitations/implications: The statistical model and conclusions drawn from it are only valid in the interior of the investigated region of the parameter space. Additionally, it might not be possible to find a local minimum of the aeration measure (damping lag) inside the selected region of the parameter space; a/the (depending on the context) global minimum located at the boundary might be the only possible solution. Practical implications: The optimal value of parameters is not unique and thus additional sub-criteria (cost/durability) can be imposed. Conducting tests in an organized manner and according to the Six Sigma methodology allows the design optimization process to be expedited and unnecessary costs to be eliminated. Originality/value: Improvements in understanding and measuring aeration effects constitute a clear foundation for further product optimization. Signal post-processing algorithms are essential for the statistical analysis and are the original contribution of this work.
Hydrological modelling is essential for improving water management and planning efficiency and sustainability. In this study, lumped conceptual models [i.e., Génie Rural à 4 paramètres Journalier (GR4J), Génie Rural à 6 paramètres Journalier (GR6J)] and wavelet-based data-driven models [Wavelet-Genetic algorithm-Artifcial neural network (WGANN), Wavelet-based support vector regression (WSVR)] were used for daily rainfall-runoff modelling by using three gauging stations, namely Çaydere Eğirdir Göl Giriş, Kargı Ç. Türkler and Naras D. Şişeler, in semi-arid and humid areas of Antalya basin, Turkey. The Nash Sutcliffe efficiency (NSE), index of agreement (d) and root mean square error (RMSE) were used to evaluate the model performance. Although conceptual and data-driven models yielded a good performance, data-driven models could be more helpful, especially in semi-arid and small basins, challenging for conceptual models due to nonlinearity and complexity. The best runoff forecasting performance improvement was observed in Çaydere Eğirdir Göl Giriş with the WGANN (NSE=0.96, d=0.99, RMSE=0.5 mm/d), WSVR (NSE=0.95, d=0.99, RMSE=0.6 mm/d) against the GR4J (NSE=0.53, d=0.79, RMSE=1.8 mm/d) and the GR6J (NSE=0.49, d=0.78, RMSE=1.8 mm/d). It was also found that the GR4J and GR6J yielded a similar performance. Data denoising via wavelet transformation and input selection had a significant role in developing performance for the data-driven models. Data-driven models yielded better results for the forecasting of extreme flows. In this regard, using and integrating the useful parts of the conceptual and data-driven models could be more favourable.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.