LOGIC BASED PATTERN DETECTION BASED ON MULTI-LEVEL PROPOSITIONAL PROCESS
Journal Title: International Journal of Computer Science & Engineering Technology - Year 2012, Vol 3, Issue 12
Abstract
Data mining is the procedure of hauling out enviable information or remarkable patterns from presented databases for precise purposes. The effectiveness of the rules produced depends on the support threshold, which consecutively involve decisions finished employing these rules. Nearly all of the earlier strategies put a distinct minimum support threshold for all the items or item sets. But in genuine applications, diverse items might have diverse criteria to review its significance. The support necessities ought to then differ with diverse items. The existing work presented a structure for discovery of patterns based on propositional logic which evaluate the coherent rules (i.e., knowledge discovery). The discovery of association rules openly from the logical rules with no minimum support threshold is evaluated. Nevertheless it processes on distinct level of propositional logic, where hierarchical schemes of the knowledge domain cannot be derived. To enhance the pattern discovery process, the proposed work extends the pattern discovery process with coherent rule generation framework in terms of multi-level hierarchical property propositions. The multi-level coherent rule structure produce rules coming from diverse levels and determine highest recurrent item sets at inferior level. The propositional logic process formed the multilevel connection rules from logical rules and utilizes bottom-up progressive extending technique. The bottom up progressive method develops the effectiveness of rules being produced devoid of minimum support threshold. Experimentation are carried out using real data set to assess multilevel association rules capably using concept hierarchies, which describes a series of mappings from a position of low level concepts to advanced level. The presentation of rule creation is measured up to that of the presented single level coherent rule miners.
Authors and Affiliations
PRASADH. K , SUTHEER. T
Keywords
Related Articles
- EP ID EP103732
- DOI -
- Views 121
- Downloads 0
PERIODICITY DETECTION ALGORITHMS IN TIME SERIES DATABASES-A SURVEY
Periodicity mining is used for predicting different applications such as prediction,forcasting etc.It has several application in Timeseries databases.Several algorithms are present for detecting the periodicity.But most...
An SMS and USSD Model for Locationbased Mobile Advertising
The use of mobile phones to deliver context specific information in the form of advertisements tailored to a user’s profile, location among other related characteristics has been on the increase in the last few years. Lo...
Energy Cost and QoS based management of Data Centers Resources in Cloud Computing
Cloud computing is the evolutionary step to transform a large part of the Information and Communication Technology industry. It is the result of the efforts to provide the opportunity to focus on hardware and software co...
Efficient Prediction of Cross-Site Scripting Web Pages using Extreme Learning Machine
Malicious code is a way of attempting to acquire sensitive information by sending malicious code to the trustworthy entity in an electronic communication. JavaScript is the most frequently used command language in the we...
Brain Cancer Classification Using GLCM Based Feature Extraction in Artificial Neural Network
Brain tumor is one of the major reasons of death among people. It is indication that the chances of survival can be greater than before if the tumor is detected correctly at its early stage. This paper classifies the typ...