In the age of pervasive internet where people are communicating, networking, buying, paying bills, managing their health and finances over the internet, where sensors and machines are tracking real-time information and communicating with each other, it is but natural that big data will be generated and analyzed for the purpose of “smart business” and “personalization”. Today storage is no longer a bottleneck and the benefit of analysis outweighs the cost of making user profiling omnipresent. However, this brings with it several privacy challenges – risk of privacy disclosure without consent, unsolicited advertising, unwanted exposure of sensitive information and unwarranted attention by malicious interests. We survey privacy risks associated with personalization in Web Search, Social Networking, Healthcare, Mobility, Wearable Technology and Internet of Things. The article reviews current privacy challenges, existing privacy preserving solutions and their limitations. We conclude with a discussion on future work in user controlled privacy preservation and selective personalization, particularly in the domain of search engines.

The analysis of text in the form of tweets, chat or posts can be an interesting as well as challenging area of research. In this paper, such an analysis provides information about the human behavior as positive, negative or neutral. For simplicity, tweets from social networking site, Twitter, are extracted for analyzing human personality. Various concepts from natural language processing, text mining and neural networks are used to establish the final outcome of the application. For analyzing text, Neural Networks are implemented which are so modeled that they predict the Human behavior as positive, negative or neutral based on extracted and preprocessed data. Using Neural Networks, the particular pattern is identified and weights are provided to words based on the extracted pattern.Neural networks have an added advantage of adaptive learning. This application can be immensely useful for politics, medical science, sports, matrimonial purposes etc.The results so obtained are quite promising.

Data mining plays an important role in collecting information to make businesses more competitive in present business world. It is seen that the location of any business outlet is a major factor of its success. Establishing different business enterprises include a detail study of localities, people's income status living in those areas, and many other non-spatial factors. This paper is one such idea to suggest those locations for entrepreneurs, based on which they can decide on the where they can setup their business outlet. The proposed algorithm makes use of Delaunay triangulation for capturing spatial proximity and Bezier curves are used to model obstacles. The algorithm is implemented as Web application, which accepts the name of a place and collects data, form clusters and show the feasible locations of the service specified, considering the geographic irregularities and man-made obstructions. In this algorithm, spatial and non-spatial data related to a location are collected and the spatial clustering algorithm is initiated which works based on the obtained data. Clusters are formed based on the unique characteristics of each location. The experimental results are conducted on many different locations of India and in this paper results are shown for three places namely, Mysuru, Patna and Mumbai. The results have shown expected and exciting results.

Predicting human behavior based on the usage of text on social networking sites can be a challenging area of interest to a particular community. Text mining being a major interest in Data Mining has vast applications in various fields. Clients can assess an individual’s behavior using the proposed framework that is based on person’s textual interaction with other people. In this paper, a framework is proposed for predicting human behavior in three phases- Text Extraction, Text cleaning and Text Analysis. For cleaning text, all the stop words have been removed and then the text is utilized for further processing. Then, the terms from the text are clustered based on semantic similarity and then gets associated with respective physiological parameters that identify a human behavior. This application is best suited for the fields of Criminal Sciences, Medical Sciences, Human Resource Department and Political Science and even for Matrimonial purposes. The proposed framework is applied on some famous world known celebrities and the results are quite encouraging.