OPTIMIZATION OF FREQUENT REPRESENTATIVE PATTERN SETS USING NAIVE BAYES TECHNIQUE
Journal Title: International Journal of Engineering Sciences & Research Technology - Year 30, Vol 5, Issue 7
Abstract
Frequent item sets plays a vital part in abundant actions of data mining which continuously endeavor to determine fascinating arrangement through number of databases like association rules, sequences, classifiers based on correlations, episodes, clusters and so on. The time period required to generate periodic item sets enacts a crucial aspect. Various algorithms are developed, taking only time aspect into account. This turns up requirement of determining small number recurrent producing patterns. Here, fundamental problems on mining of periodic item sets as well as pattern sets are elaborated. Frequent pattern mining generally develops a large number of recurrent patterns that depict complex concerns on understanding, viewing and next investigation of derived patterns. This results in determining a small number of classical arrangements of patterns to approximate all other patterns to the best suitable extent. In latest practices in the research of recurrent mining of pattern to determine a minimal classical set of pattern having zero error algorithm known as MinRPset is implemented. MinRPset determines modest result which one can probably implement in experimental procedures for the mentioned situation and it takes ample time to conclude as soon as the count of periodic closed arrangement patterns is less than one million. MinRPset is highly utilizes memory space and time on numerous heavy datasets when the count of recurrent closed pattern is greater. To solve this difficulty, another algorithm called FlexRPset is utilized, that introduces one supplementary parameter K for facilitating users for making adjustment amongst efficiency and result size. An additional view to provide users to make intervene satisfaction is utilized. Some recurrent pattern mining usually develops a more count of periodic patterns that require a huge objective of understanding, visualizing and next investigation of the developed patterns. This increases the demand to find less count of repeatedly occurring patterns. In the proposed work, system classifies the patterns on using Machine Learning, which classifies the items browsed by user for finding exact pattern set. By using naive bayes technique, service provider can increase his sell and also the customers can be advised for exact items they need through exact pattern generation.
Authors and Affiliations
Preeti C. Tarange*
Keywords
Related Articles
- EP ID EP159530
- DOI 10.5281/zenodo.58653
- Views 83
- Downloads 0
EXPERIMENTAL STUDY ON BEHAVIOUR OF FIBER REINFORCED TERNARY BLENDED CONCRETE USING DIFFERENT TESTS
The construction techniques have been modernized with focus on high strength, dense and uniform surface texture, more reliable quality, improved durability, and faster construction. High strength concrete is a typ...
MULTI-MODEL BIOMETRICS AUTHENTICATION FRAMEWORK
Authentication is the process to conform the truth of an attribute claimed by real entity. Biometric technology is widely useful for the process of authentication. Today, biometric is becoming a key aspect in a multitud...
OPTIMAL LOCATION OF TCSC AND SVC DEVICES TO ENHANCE POWER SYSTEM SECURITY
Contingency analysis in power system security is an important tasks in emerging energy management systems. Contingency ranking using the performance index value is a method for the line outages in a system, which...
MOSAICING IN VARIOUS IMAGES: A SURVEY
K-means clustering and watershed are the key techniques used as a combination, Edge detection procedure and region, merging are also used. Merging method based on mean gray values and edge strengths (T1, T2) are u...
Study on Design Parameters Effecting Mechanical Carrot Harvester
The design parameters of a mechanical carrot harvester were determined by conducting experiments on a test set-up having provision to vary design variables. Three lengths of soil separators (40, 60 and 80 cm), th...