Web Usage mining consists of three main steps: Pre-processing, Knowledge Discovery and Pattern Analysis. The information gained from the analysis can then be used by the website administrators for efficient administration and personalization of their websites and thus the specific needs of specific communities of users can be fulfilled and profit can be increased. Also, Web Usage Mining uncovers the hidden patterns underlying the Web Log Data. These patterns represent user browsing behaviours which can be employed in detecting deviations in user browsing behaviour in web based banking and other applications where data privacy and security is of utmost importance. Proposed work pre-process, discovers and analyses the Web Log Data of Dr. T.M.A.PAI polytechnic website. A neuro-fuzzy based hybrid model is employed for Knowledge Discovery from web logs

In today's era, millions of users make use of internet in their day to day activities. In various commercial and non-commercial applications it is very essential to protect the confidentiality of data. Hence it is mandatory to have automated tools to protect the data. With the introduction of distributed computing systems, network communication security has become another major aspect. All academic institutions, business and government organisations make use of distribution networks to process their data. Therefore, protecting data during a transmission within a network of computers located at different geographical locations where end users working simultaneously are inevitable. Also for people making modern day e-transactions like money transfer, online shopping etc., a secure framework is adequate. Cryptography and Steganography are a few of the methods used for secure communication. In this paper, a technique has been proposed for providing multilayer security by combining cryptography and steganography. The proposed RSA cryptographic algorithm integrated with dual audio steganography algorithm provides a higher level of security.

Titanic disaster occurred about 100 years back but still it attracts the researchers to understand and study that how some passengers survived and others perished. In this work, the characteristics of the passengers will be identified and the relationship of survival chance from the disaster is found. Feature engineering techniques will be performed, where the alphabetic values will be changed to numeric values, the family size will be calculated. Also, we will extract the title from the name, and deck label from ticket number. Classification is done using Decision tree machine learning classification algorithm using two classes which are survived and not survived. R programming has been used for its implementation. Clustering is performed using KMeans machine learning algorithm. Its implementation has been done using Python programming.

Tele-radiology is a technology that helps in bringing the communication between the radiologist, patients and healthcare units situated at distant places. This involves exchange of medical centric data. The medical data may be stored as Electronic Health Records (EHR). These EHRs contain X-Rays, CT scans, MRI reports. Hundreds of scans across multiple radiology centers lead to medical big data (MBD). Healthcare Cloud can be used to handle MBD. Since lack of security to EHRs can cause havoc in medical IT, healthcare cloud must be secure. It should ensure secure sharing and storage of EHRs. This paper proposes the application of decoy technique to provide security to EHRs. The EHRs have the risk of internal attacks and external intrusion. This work addresses and handles internal attacks. It also involves study on honey-pots and intrusion detection techniques. Further it identifies the possibility of an intrusion and alerts the administrator. Also the details of intrusions are logged.

India is a nation where more focus is now towards the field of digitalization. Government is focusing on the initiative “Digital India” where it is more encouraged to make use of online facilities for various types of money transfers. People are encouraged to make use of NEFT, RTGS and various other modes in which the money is transferred directly to the other persons account rather than the liquid cash being circulated. Internet banking is one of the ways where in money transfer is usually done by the user instead of stepping into bank. Here user needs to remember all the credentials to login. Another approach is proposed in this paper where biometric information of the user can be used for login process. All the bank accounts are now linked with the Aadhaar number of the person. Aadhaar number is a unique identification number given to all the Indians which uniquely identifies the person. Aadhaar will have all the details of a person's biometric information like finger print which is used for the login process of the internet banking. This eases the login process of the internet banking. The main focus here is to make the personal banking experience of the user better.

The image processing task for conversion of image style is a tedious task. The previous approaches carried out in this area lacked in extracting higher level of abstractions from the input image, hence failed to generate image content from the acquired knowledge. Hence to achieve higher level knowledge from the object to be recognized, we make use of convolutional neural networks which aid in achieving higher quality images. The proposed work uses a Neural Algorithm that produces an image with a new style using the knowledge acquired from input image. Our work provides a new approach for deep representations of images using Neural Networks, hence demonstrating the generation of high level image synthesis and manipulation.