Manuscripts

Recent Papers

Research Paper

Review on credit card fraud detection and classification by Machine Learning and Data Mining approaches

The strategies for this are divided into 2 broad types: fraud detection as well as consumer activity analysis. The initial category of strategies works with controlled recognition processes at transaction stage. Transactions are classified as illegitimate or regular depending on preceding historical evidence in such systems. This dataset can then be used to construct classified models that can forecast the status of new documents (normal or fraudulent). A standard two-classification function, including rule inference, decision trees, as well as neural networks, has various model development approaches. This method has been shown to accurately identify most previously found fraud techniques, often known as identification of misuse essential to illustrate the main discrepancies in an overview of consumer behaviour and methods to fraud investigation. The system of fraud detection can identify established tricks from fraud, with a small false positive rate. Such schemes derive the sign as well as pattern of fraudulent strategies provided in the revelation data set as well as can therefore quickly decide precisely that frauds; the machine is witnessing at the moment.

Published by: Aaushi Sharma, Neha Bathla

Author: Aaushi Sharma

Paper ID: V6I4-1419

Paper Status: published

Published: August 20, 2020

Full Details
Research Paper

A study on influence of store ambience on consumers purchase behaviour

Store ambiance is referred to as a store's physical characteristics which are used to develop a retail unit image and gaining customers. The project is to study the influence of the store ambiance on consumer purchase behavior. It includes factors such as storefront, marquee, entrance, display, flooring, lighting, colors, fragrance, music, fixtures, and so on. Some factors majorly affect consumer purchase behavior they are lighting, fragrance, music, window display, color co-ordination, product set-up. Therefore by analyzing the already existing ambiance of a retail store, a survey is conducted to know the consumer's preferences on store ambiance that affects their purchase and then concluded based on the results obtained.

Published by: S. Ilakya , S. Sandhya, Dr. D. Vijayalakshmi

Author: S. Ilakya

Paper ID: V6I4-1397

Paper Status: published

Published: August 20, 2020

Full Details
Case Study

Over-dues forecasting using ARIMA Technique

The work presented in this paper establishes an enrichment in modeling and forecasting over-dues for Beverages manufacturing company. A time-series modeling technique used to forecast over-dues for ABinBEV (Beer manufacturing company). Our work demonstrates how historical over-dues data utilized to predict future over-dues. The historical over-dues information used to develop several Autoregressive Integrated Moving Average (ARIMA) models by using Root mean squared error (RMSE) and the most suitable ARIMA model found to be ARIMA (2, 1, 0). and validation performed by comparing the accuracy of the models with three types of accuracy criteria, which are Mean square error (MSE), Root Mean Squared Error (RMSE), and Mean absolute error (MAE).

Published by: A. Kalyan Aravind Kumar

Author: A. Kalyan Aravind Kumar

Paper ID: V6I4-1392

Paper Status: published

Published: August 20, 2020

Full Details
Research Paper

Design and analysis of cocoon extractor

This paper tells us about Design and Analysis of Cocoon Extractor machine. One of the traditional methods for growing cocoon is by using chandrika. This method is widely used in SOUTH INDIA. The removal of cocoon from chandrika is art and it’s a careful practice followed by farmers. During extraction process of cocoon farmers face many problems i.e., labour problem, wearing of bamboo strips due to continuous strips, pricking of bamboo strips into fingers, time management and many such problems. The major problem we are facing while building the machine is that the texture of cocoon is too delicate. To solve these problems, we decided to use air as medium to extract cocoon from chandrika. This can be achieved by vacuum pressure. The basic principle we are using in this machine is vacuum cleaner’s principle. After survey we found that we need high suction pressure to suck the cocoon from the chandrika. We will be using centrifugal impeller to create high suction pressure which is driven by universal DC motor.

Published by: Yeshwant J., Samarth B. Deshpande, Yashvanth Naik M. M., Sadan Gowda V., Manjunath Naik H. R.

Author: Yeshwant J.

Paper ID: V6I4-1410

Paper Status: published

Published: August 18, 2020

Full Details
Review Paper

Polymer-concrete composites: Impregnation

We see monuments and beautiful old buildings around us that were built ages before. They not only provide great visual joy but also remind us of our history and connects to our ancestors. With time, several of these have deteriorated causing the risk of demolition at any time. Clearly these buildings require reinforcement for sustainment. Advancement in concrete technology has led to polymer impregnated concrete which can be used for the restoration of such structures.

Published by: Ankur Kapooria

Author: Ankur Kapooria

Paper ID: V6I4-1404

Paper Status: published

Published: August 18, 2020

Full Details
Technical Notes

Accuracy of Machine Learning Models for Plant Disease Detection

Agriculture is the backbone of a nation. India has about 96 million hectare of irrigated land. With the amount of land that is cultivated as farmland, detection and prevention of diseases in crops is paramount. When diseases affect plants, particularly through their leaves it effects the production of agricultural produce and decreases profitability of a given crop. Timely identification of these diseases is very challenging in affected plants. A reliable and fast way for the detection of diseases is necessary. Detecting disease may be a key to stop agricultural losses. The aim of this is to develop a software system that is able to efficiently find and classify diseases occurring in plants. The pictures of leaves can be used for detecting the plant diseases. Therefore, use of image process technique to find and classify diseases in agricultural applications is useful.

Published by: Harshith P. K., Bitopan Deka, Nikhil N., Sumanth T. S.

Author: Harshith P. K.

Paper ID: V6I4-1396

Paper Status: published

Published: August 18, 2020

Full Details