In Machine Learning, Decision tree is the mostly used classifier for predictive Modeling. The C4.... more In Machine Learning, Decision tree is the mostly used classifier for predictive Modeling. The C4.5 classifier suffers from overfitting; poor attribute split technique, inability to handle continuous valued and missing valued attributes with high learning cost. Among all, overfitting and split attribute has high impact on the accuracies of prediction. The Efficient Back-track pruning algorithm is introduced here to overcome the drawback of overfitting. The proposed concept is implemented and evaluated with the UCI Machine Learning Hungarian database. This database having 294 records with fourteen attributes were used for forecasting the heart disease and relevant accuracies were measured. This implementation shows that the proposed Back-track pruned algorithm is efficient when compared with existing C4.5 algorithm, which is more suitable for the application of large amounts of healthcare data. Its accuracy has been greatly improved in line with the practical Health care Historical da...
Techniques Tanimoto correlated feature selection system and hybridization of clustering and boosting ensemble classification of remote sensed big data for weather forecasting
International Journal of Recent Technology and Engineering
Although machine learning has long provided a powerful approach to prediction, its applicability ... more Although machine learning has long provided a powerful approach to prediction, its applicability has been somewhat emerging right now because of the large requirements in the various field. In recent years a number of new predictions with greatly reduced algorithm requirements have been developed. The purpose of this paper is to survey the various techniques that using now a days. The approaches of machine learning and the algorithms are included in this review. Several applications of the new techniques are discussed.
An Investigation Study on Clustering and Classification Techniques for Weather Forecasting
Journal of Computational and Theoretical Nanoscience
Weather forecasting is the prediction of atmosphere state for particular location by using princi... more Weather forecasting is the prediction of atmosphere state for particular location by using principles of physics provided by many statistical and empirical techniques. Weather forecasts are frequently made by collecting quantitative data about current state of atmosphere through scientific understanding of atmospheric processes to illustrate how atmosphere changes in future. Current weather conditions are collected through the observation from the ground, ships, aircraft, radio sounds and satellites. The information is transmitted to the meteorological centers where the data are collected and examined for prediction. There are diverse techniques included in weather forecasting, from relatively simple observation of sky to complex computerized mathematical models. But, the existing techniques failed to predict the weather with higher accuracy and lesser time. In order to improve the prediction performance, the machine learning and ensemble techniques are introduced.
Banking systems collect huge amounts of data on day to day basis, be it customer information, tra... more Banking systems collect huge amounts of data on day to day basis, be it customer information, transaction details, risk profiles, credit card details, limit and collateral details, compliance and Anti Money Laundering (AML) related information, trade finance data, SWIFT and telex messages. Thousands of decisions are taken in a bank daily. These decisions include credit decisions, default decisions, relationship start up, investment decisions, AML and Illegal financing related. One needs to depend on various reports and drill down tools provided by the banking systems to arrive at these critical decisions. But this is a manual process and is error prone and time consuming due to large volume of transactional and historical data. Interesting patterns and knowledge can be mined from this huge volume of data that in turn can be used for this decision making process. This article explores and reviews various data mining techniques that can be applied in banking areas. It provides an overview of data mining techniques and procedures. It also provides an insight into how these techniques can be used in banking areas to make the decision making process easier and productive.
International Journal of Intelligent Engineering and Systems, 2017
In the field of weather forecasting, especially in rainfall prediction many researchers employed ... more In the field of weather forecasting, especially in rainfall prediction many researchers employed different data mining techniques. There is numerous method of organizing agricultural engineering substance and it remains an open research issue particularly when taking to distinctive arrangements of clients-farmers, agricultural engineers, agri-organizations-both from proficiency point of view. Keeping these factors Indian farmers in mind, we have chosen to do research on efficient dissemination of rainfall forecasting to safeguard farmers from crop failure using optimized neural network (NN) model. Here, at first, we generate the feature matrix based on five feature indicator. Once the feature matrix is formed, the prediction is done based on the hybrid classifier. In hybrid classifier, particle swarm optimization algorithm is combined with Grey Wolf optimization for training the RBF NN. The performance of the algorithm is analyzed with the help of real datasets gathered from pechiparai and perunchani regions.
Journal of the Association of Arab Universities for Basic and Applied Sciences, 2016
This paper proposes to evaluate the adaptability risk in money laundering using Bitmap Index-base... more This paper proposes to evaluate the adaptability risk in money laundering using Bitmap Index-based Decision Tree (BIDT) technique. Initially, the Bitmap Index-based Decision Tree learning is used to induce the knowledge tree which helps to determine a company's money laundering risk and improve scalability. A bitmap index in BIDT is used to effectively access large banking databases. In a BIDT bitmap index, account in a table is numbered in sequence with each key value, account number and a bitmap (array of bytes) used instead of a list of row ids. Subsequently, BIDT algorithm uses the ''select" query performance to apply count and bit-wise logical operations on AND. Query result coincides exactly to build a decision tree and more precisely to evaluate the adaptability risk in the money laundering operation. For the root node, the main account of the decision tree, the population frequencies are obtained by simply counting the total number of ''1" in the bitmaps constructed on the attribute to predict money laundering and evaluate the risk factor rate. The experiment is conducted on factors such as regulatory risk rate, false positive rate, and risk identification time.
Software-Architecture for Object Oriented Systems- Usability Patterns
Abstract Over the years the software engineering community has increasingly realized the importan... more Abstract Over the years the software engineering community has increasingly realized the important role of software architecture plays in fulfilling the quality requirements of a software system. It has been experienced that Software Architecture (SA) constrains the achievement of ...
Abstract The software engineering projects [22, 23] reveals that a large number of usability rela... more Abstract The software engineering projects [22, 23] reveals that a large number of usability related change requests are made after its deployment. Fixing usability problems during the later stages of development often proves to be costly, since many of the necessary changes require ...
The banking industry has undergone various changes in the way they conduct the business and focus... more The banking industry has undergone various changes in the way they conduct the business and focus on modern technologies to compete the market. The banking industry has started realizing the importance of creating the knowledge base and its utilization for the benefits of the bank in the area of strategic planning to survive in the competitive market. In the modern era, the technologies are advanced and it facilitates to generate, capture and store data are increased enormously. Data is the most valuable asset, especially in financial industries. The value of this asset can be evaluated only if the organization can extract the valuable knowledge hidden in raw data. The increase in the huge volume of data as a part of day to day operations and through other internal and external sources, forces information technology industries to use technologies like data mining to transform knowledge from data. Data mining technology provides the facility to access the right information at the right time from huge volumes of raw data. Banking industries adopt the data mining technologies in various areas especially in customer segmentation and profitability, Predictions on Prices/Values of different investment products, money market business, fraudulent transaction detections, risk predictions, default prediction on pricing. It is a valuable tool which identifies potentially useful information from large amount of data, from which organization can gain a clear advantage over its competitors. This study shows the significance of data mining technologies and its advantages in the banking and financial sectors.
Uploads
Papers by Siva Balan