The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
In this paper, a software toolkit has been developed for in silica prediction of the differentiation destiny of Mesenchymal Stem Cells (MSCs) in vitro. The software toolkit was developed in CLIPS (C Language Integrated Production System) as an expert system, with a java-based GUI. This toolkit utilizes the rules obtained from previous experimental data via data mining techniques, based on which the...
For the problem of feature selection of stock, this paper presents a new algorithm which is the optimal combination of Principle Components Analysis (PCA) with Support Vector Machines (SVMs). The new algorithm is based on weight measure. Because of specialty of this problem, a weight measure is learned by PCA and SVMs with linear kernel function. Good stock and bad stock with many features belong...
Attribute importance ranking still is a key problem required to be solved for the classification problem. The lack of efficient heuristic information is the fundamental reason that affects the attribute selection in data mining. In this paper, to determining the importance level of the attributes, a new measure based on partial derivative distribution of the classification hypersurface output corresponding...
This paper uses GMDH method to establish a prediction model to forecast the vehicles for business transport of Guangxi in China, since the original samples of the output value of transport & storage of Guangdong are less enough to be used with the traditional methods. Compared with traditional linear regression and artificial neural network, the predicted results show that GMDH method is an effective...
In order to increase the protection ability of the network intrusion diction system (NIDS), it is important to gather the host information of the intruder. In the proposed IDS called NIDS-SA, three basic components are developed to support the active monitoring capability, Intrusion Detection Node (IDN), Intrusion Detection Coordinator (IDC), and Snooper Agent (SA). The IDN is used to capture packets,...
A computer aided system is proposed to screen and sort the non-patent literatures in this paper. It is introduced that the present situation and strategy of screening in data processing. The system here is a man-machine interactive system. First, knowledge and experience which stand for the know-how of the experts were inserted into the knowledge base, including the screening rules and principle in...
The research of distinction of name ambiguity in the field of information retrieval could enhance searching effect. Therefore, it plays an important role to mine the data of name ambiguity in order to obtain useful knowledge. In this paper, we focus on the problem of traditional evaluation and ranking method used in the clustering. Traditional evaluation and ranking method ignores the association...
The basic philosophy of mining relevance rule algorithm in knowledge discovery is introduced. After analysis on obvious shortcoming of frequent Pattern tree (FP-tree) algorithm in relevance rule mining, an improved algorithm of FP - tree growth is proposed to enhance the performance of mining relevance rules based on mapping construction of FP-tree. The effectiveness of the method proposed here is...
Most previous approaches to automatic audio events (AEs) annotation are based on supervised learning which relies on the availability of a labeled corpus to train classification models. However, instance annotation is often difficult, expensive, and time consuming. In this paper, we apply semi-supervised learning with transductive Support Vector Machine (TSVM) algorithm to automatic AEs annotation...
To solve the date updates problem of drug Data Warehouse(DW), the General methods of data updates in DW and the problems that may exist in the process of data extraction, transmission and uploading were analyzed and studied. To achieve the requirements of reliability and rapidity of drug data updates, a method of incremental updates with triggers and intermediate table was proposed, and the design...
According to the theory of power load forecasting, data mining based on historical data of power load is used in load predicting. During the practical application, there are some errors in the data collection, and a load forecasting curve often contains big jagged edges. This paper presents a new outlier data mining approach. It finds sharp angle points between two lines, which correspond to outliers...
Through the wide use of E-commerce, the acquisition of personalized need is key to effective recommender. From the view of knowledge acquiring, case intelligence is a comprehensive expression which is integrated representation of human sense, logics and creativity, and can acquire the user's preferences from the former stored cases. As the E-commerce is under much complex conditions, this paper presents...
In this paper, Computer Generated Forces (CGF) behavior modeling was studied from the viewpoint of associate data mining, for the large quantity of data, rules and models in its process. Because CGF behavior models data source was the combination of staticDB and dynamic data stream, the paper advanced the methods of item truncation and aim-pattern restriction. Through pretreatment, coding, searching...
It is urgent to solve the problem of how to accurately understand users' behaviors of visiting websites in the development of e-commerce. Web log mining is an important research method in addressing the problem. In this paper, we propose the new concept of interest pheromone, and on the basis of which design a group users' navigation path mining algorithm based on ant colony algorithm. The experimental...
Rule Induction and finding out potential relations between indicator value and evaluation result are the bottle neck problem in comprehensive evaluation. Rule induction based on Classification Consistency Rate (RICCR) can be applied to solve the problem. The application of RICCR will help decision makers to obtain concise rules from databases based on rough set, both flexible and generalized. This...
Uncertainty and imprecision problem is a common phenomenon in the medical diagnosis, this paper present a novel method to deal with it. The first, some new concepts such as elastic trust, elastic trust granular (ETG), inclusion and similarity relation of elastic trust (ISRET), and so on are introduced. The second, on that basis, the paper puts forward a novel distance of elastic trust granular, such...
Customer relationship management competition plays a very important role in modern enterprises. Good relations between customers and companies are important aspects for enterprise's survive and development. In this paper, through combing customer relationship management research results from home and abroad and our own study on the ceramics production process, we come up with “ceramic enterprise customer...
In this thesis, we will introduce the concepts of data mining technology and customer relationship management to analyze the advantages and disadvantages of decision tree and neural network. With the decision tree and neural network fusion algorithm, we shall find its necessity in bank-customers management system application in the banking sector development and will explain the detailed applications...
Text classification is an important research field of data mining topics. This article brings a mutual information and information entropy pair based feature selection method (MIIEP_FS) based on the theory of information entropy and information entropy pair concept. This method measure the classification effect using feature by mutual information method and show the difference extent between the features...
Tax information system has been developed for many years. Most of the tax management systems that are using now belong to services-faced system. After being used many years, there accumulated a lot of historical data, but most of this data are decentral, static and low aposteriori. This data hasn't been filtered, abstracted, analyzed and predict. They can't supply information support to the leader...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.