Eliminating the redundant data for side channel in cloud
In recent years, the amount of data stored at the cloud storage is increasing rapidly due to the prevalence of data outsourcing. In order to be cost-effective and to reduce the bandwidth consumption, cloud stores use cross-user client-side data deduplication which eliminates the need to store redundant copies by keeping only a single copy of the data at the cloud storage. Cloud storage providers generally adopt data deduplication, a technique for eliminating redundant data by keeping only a single copy of a file, thus saving a considerable amount of storage and bandwidth. Even then, the attackers use many deduplication protocols to steal information. For example, an attacker can perform the duplicate check to verify whether a file (e.g., a pay slip, with a specific name and salary amount) is already stored (by someone else), hence breaching the user privacy.
Published by: C. V. Usha Kiran, C.Reddineelima, B. Lijitha, K.Somesh, G.Venugopal
Author: C. V. Usha Kiran
Paper ID: V5I2-1665
Paper Status: published
Published: April 1, 2019
Analysis and prevention of suicides using World Health Organization (WHO) data
Youth committing suicides is the major public health issue in the world. It is the third leading death reason in the United States of America for ages between 13-18. Many adults who have suicidal thoughts or make a decision of committing suicide plan never approach any professional for help or advise. Most of the cases, people attempting suicide don’t express it a lot in person due to the judgmental attitude of society. We have come with an idea of using WHO published data on suicide. With this we evaluate results to recognize and reduce the suicidal risk of the age group committing more suicides. We can use the same data to target the right audience through social media and campaigns to prevent it and reduce the cases. Linear regression and Random Forest algorithms are used which can predict the possible cases of suicides. We included a prediction algorithm called fbprophet. Once the analysis of the dataset is completed, we visualize the data and the group of audience committing more suicides will be targeted on social media. In the part of prevention, the audience is targeted on Facebook by using a parameter like the age, sex, and region. For this Facebook provides a paid medium through which we can provide awareness and motivation for targeted people.
Published by: Ronanki Bharath, Janaki S., Bhanu Narra, Rishabh Gupta, S. Rama
Author: Ronanki Bharath
Paper ID: V5I2-1672
Paper Status: published
Published: April 1, 2019
Optimization of generation-specific buying plan for premium lingerie brand
Buying a plan is the most important aspect of a profitable retail business. Retail buying plan includes sales forecast process by analyzing the factors affecting it namely aggregation of sales and exogenous variables. The conventional buying plan is centered around a common customer profile. However, in the future, we need a generation specific buying plan catering to their specific design expectations. In this study propose a new replenishment strategy formulated for Gen Z customers. The new buying plan strategy weighs the recent trends and prevailing sales scenarios to forecast the replenishment quantity and re-order point. The evolved strategy carries its references to sales history data, trends, aggregation of data for the product topology, especially for Gen Z customers.
Published by: Vinosha B., Nithyaprakash V.
Author: Vinosha B.
Paper ID: V5I2-1684
Paper Status: published
Published: April 1, 2019
Decentralized secure money using blockchain
The blockchain is a decentralized transaction and managing the data. This technology introduced on the first for Bitcoin cryptocurrency which makes more easier for the transaction. Blockchain technology has highly raised up in 2008. The main purpose of the Blockchain technology is to provide security, anonymity and data integrity without any third party organization in control of the transactions. This helps on improving in research areas especially from the perspective of technical challenges and limitations. In this research, we have conducted survive mapping study with the goal of presenting the Blockchain technology. The main objective of the researchers is to make the challenges in understanding the current research areas. This the future way to developing more and more on Blockchain technology. A blockchain is just a chain and list of blocks which secures the transfer of funds, by using a digital signature algorithm to prove ownership. Each block in the blockchain will have its own digital signature. We create a web page that contains the login and registration page. We create a key pair of each user that contains the private key and public key. Every time we check this blockchain is valid or not. The main reason for the paper is to research and focus on revealing and improving limitations of Blockchain from privacy and security perspectives. On the basis of this studying the blockchain we highly recommend the researchers to develop in the future.
Published by: Dr. S. Gunasekaran, Dr. G. Ravi Kumar, S. Nirmal Balaji, Gokul R., R. Mahilan, Pooja M.
Author: Dr. S. Gunasekaran
Paper ID: V5I2-1695
Paper Status: published
Published: April 1, 2019
Automotive control system using CAN protocol
In the latest years, various developments are done in Automobiles Industries. This automobile is being developed by more of electrical parts for efficient operation and development of CAN based automobile control system CAN is a standard protocol for efficient and reliable communication between the sensor, actuator, controller and other nodes in real-time applications. CAN enable us to connect all the devices together using only two wires. The efficient data transfer rate up to 1Mbps. The CAN Protocol is used to exchange information between ECU. To provide an accurate result to the driver different sensors are used like Temperature sensor, Fuel level sensor, RPM sensor Ultrasonic Sensors, etc. CAN bus to implement the system to alert the driver. This paper provides the development and implementation of various previously presented systems, which are useful to improve the driver vehicle interface.
Published by: Surekha P. Gaikwad, Atmeshkumar S. Patel
Author: Surekha P. Gaikwad
Paper ID: V5I2-1652
Paper Status: published
Published: April 1, 2019
Prognostication of breast cancer using data mining and machine learning
In this study, WPBC that is Wisconsin Prognostic Breast Cancer (original) dataset to find an efficient predictor algorithm, to predict the recurring and non-recurring nature of breast cancer. Breast cancer is the most common disease found among the women, it is difficult for the physicians to know the exact reason behind breast cancer, and they need a smart system for predicting the illness on time before it is too late to be treated. It is one of the crucial reasons for death among females all over the world. The cancer tumor is generally categorized into benign and malignant tumors. Using machine learning and data mining techniques it can easily identify the cancer cells, provide benefits to the medical system. Use of classification algorithm C5.0 and XGBOOST sums up to have a considerably better result, improving the performance of the system. İn Classification, the calculation separates the information into unmistakable gatherings. It fundamentally pursues two stages of preparing first to learn and after that to a group.
Published by: Shriya D. Narvekar, Ameya Patil, Jagruti Patil, Saniket Kudoo
Author: Shriya D. Narvekar
Paper ID: V5I2-1658
Paper Status: published
Published: April 1, 2019
