International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 06 | Jun 2023
p-ISSN: 2395-0072
www.irjet.net
Classification and Prediction Based Data Mining Algorithm in Weka Tool Renu1, Kanika2 ------------------------------------------------------------------------***-----------------------------------------------------------------------
Abstract-Process of extract unseen and hidden
information from large set of data is Data Mining. Different techniques and algorithm are used to get the meaningful information from the large set of data. Different classification algorithm are used just like J48, SMO, REP tree, Naïve Bayes, Multilayer perception to extract meaning information from large set of dataset. Predictive data mining that use historical data, statistical modeling, data mining technique and machine learning to make prediction about future outcomes. Predictive analytics used in different area to identify risks and opportunities. Weka tool are use to predict new data using classification and different classifier J48,SMO,REPTree,Naïve Bayes, Multilayer Perception are classify with dataset and find accuracy of Multilayer perception is more efficient in accuracy.
2. Neural network. 3. Rule Induction. Regression. Prescriptive Modeling Pattern Mining. Anomaly Detection.
3. Methodology Weka contains a collection of classifier for data analysis with graphical user interface for easy access. Original non-Java version of weka was a Tel/TK front-end to modeling algorithms implemented in other programming languages plus data preprocessing utilities in C and a make file based system.Orignal version was design as a tool for analyzing data from agriculture domains. Weka3 java based version developed in 1997 is used in different application areas particularly for education purposes and research. Several standard data mining tasks data preprocessing, clustering, classification, regression, visualization and feature selection supported by weka.Input to weka is expected to be formatted according to the attributed relational file format.
Keywords: Data mining, Weka tool, J48 algorithm classification, Naïve Bayes
1. Introduction Huge amount of data is collected daily in this information era. Analyzing huge amount of data and extract information from that data is necessity to achieve goals. In data mining data cleaning, incorporating earlier knowledge on data set and interpreting perfect solution from the pragmatic results. Data mining[1] tool weka use to predict new data using selling house dataset. Efficiency of different classifier is calculated using confusion matrix and finds multilayer perception classifier has higher accuracy.
2. Related Technique in data mining Different data mining techniques [3] to extract insights in data but type of data mining technique used depends on their data and goals. To extract information from data a wide variety of data mining technique are employed.
Figure 1 Weka Data Mining Tool
Descriptive Modeling Clustering Association Sequential Analysis.
4.
Collection of related items of related data accessed individually is dataset. Process of preparing the raw data and making it suitable for a machine learning model just like apply filter and convert file into arff, handling missing data etc is data preprocessing. Used data in the paper is collected from kaggle.com.
Predictive Data mining Technique
Classification 1. Decision Tree
© 2023, IRJET
|
Impact Factor value: 8.226
Collect Dataset and preprocessing
|
ISO 9001:2008 Certified Journal
|
Page 908