Success of web 2.0 brings online information overload. An exponential growth of customers, services and online information has been observed in last decade. It yields big data investigation problem for service recommendation system. Traditional recommender systems often put up with scalability, lack of security and efficiency problems. Users preferences are almost ignored. So, the requirement of robust  ecommendation system is enhanced now a days. In this paper, we present review based service recommendation to dynamically recommend services to the users. Keywords are extracted from passive users
reviews and a rating value is given to every new keyword observed in the dataset. Sentiment analysis is performed on these rating values and top-k services recommendation list is provided to users. To make the system more effective and robust hadoop framework is used

Road and traffic accidents are uncertain and unpredictable incidents and their analysis requires the knowledge of the factors affecting them. Road and traffic accidents are defined by a set of variables which are mostly of discrete nature. The major problem in the analysis of accident data is its heterogeneous nature [1]. Thus heterogeneity must be considered during analysis of the data otherwise, some relationship between the data may remain hidden. Although, researchers used segmentation of the data to reduce this heterogeneity using some measures such as expert knowledge, but there is no guarantee
that this will lead to an optimal segmentation which consists of homogeneous groups of road accidents [2]. Therefore, cluster analysis can assist the segmentation of road accidents.

The idea that the purely phenomenological knowledge that we can extract by analysing large amounts of data can be useful in healthcare seems to contradict the desire of VPH researchers to build detailed mechanistic models for individual patients. But in practice no model is ever entirely phenomenological or entirely mechanistic. We propose in this position paper that big data analytics can be successfully combined with VPH technologies to produce robust and effective in silico medicine solutions. In order to do this, big data technologies must be further developed to cope with some specific requirements that emerge from this application. Such requirements are: working with sensitive data; analytics of complex and heterogeneous data spaces, including non-textual information; distributed data management under security and performance constraints; specialised analytics to integrate bioinformatics and systems biology information with clinical observations at tissue, organ and organisms scales; and

specialised analytics to define the “physiological envelope” during the daily life of each patient. These domain-specific requirements suggest a need for targeted funding, in which big data technologies for in silico medicine becomes the research priority.

Big data takes many forms, including messages in social networks, data collected from various sensors, captured videos, and so on. Big data applications aim to collect and analyze large amounts of data, and efficiently extract valuable information from the data. A recent report shows that the amount of data on the Internet is about 500 billion GB. With the fast increase of mobile devices that can perform sensing and access the Internet, large amounts of data are generated daily.

In general, big data has three features: large volume, high velocity and large variety [1]. The International Data Corporation (IDC) predicted that the total amount of data generated in 2020 globally will be about 35 ZB. Facebook needs to process about 1.3 million TB of data each month. Many new data are generated at high velocity. For example, more than 2 million emails are sent over the Internet every second.

The digitalization of mental health records and psychotherapy notes has made individual mental health data more readily accessible to a wide range of users including patients, psychiatrists, researchers, statisticians, and data scientists. However, increased accessibility of highly sensitive mental records threatens the privacy and confidentiality of psychiatric patients. The objective of this study is to examine privacy concerns in mental health research and develop a privacy preserving data analysis approach to address these concerns. In this paper, we demonstrate the key inadequacies of the existing privacy protection approaches applicable to use of mental health records and psychotherapy notes in recordsbased research. We then develop a privacy-preserving data analysis approach that enables researchers to protect the privacy of people with mental illness once granted access to mental health records. Furthermore, we choose a demonstration project to show the use of the proposed approach. This paper concludes by suggesting practical implications for mental health researchers and future research in the field of privacy-preserving data analytics.

In recent years, Big Data has changed how we do computing. Even though we have large scale infrastructure such as Cloud computing and several platforms such as Hadoop available to process the workloads, with Big Data there is a high level of uncertainty that has been introduced in how an application processes the data. Data in general comes in different formats, at different speed and at different volume. Processing consists of not just one application but several applications combined to form a workflow to achieve a certain goal. With data variation and at different speed, applications execution and resource needs will also vary at runtime. These are called dynamic workflows. One can say that we can just throw more and more resources during runtime. However this is not an effective way as it can lead to, in the best case, resource wastage or monetary loss and in the worst case, delivery of outcomes much later than when it is required. Thus, scheduling algorithms play an important role in efficient execution of dynamic workflow applications. In this paper, we evaluate several most commonly used workflow scheduling algorithms to understand which algorithm will be the best for the efficient execution of dynamic workflows.

The unprecedented growth in mobile device adoption and the rapid advancement of mobile technologies & wireless networks have created new opportunities in mobile marketing and adverting. The opportunities for Mobile Marketers and Advertisers include real-time customer engagement, improve customer experience, build brand loyalty, increase revenues, and drive customer satisfaction. The challenges, however, for the Marketers and Advertisers include how to analyze troves of data that mobile devices emit and how to derive customer engagement insights from the mobile data. This research paper addresses the challenge by developing Big Data Mobile Marketing analytics and advertising recommendation framework. The proposed framework supports both offline and online advertising operations in which the selected analytics techniques are used to provide advertising recommendations based on collected Big Data on mobile user's profiles, access behaviors, and mobility patterns. The paper presents prototyping solution design as well as its application and certain experimental results.

In the era of big data, many users and companies start to move their data to cloud storage to simplify data management and reduce data maintenance cost. However, security and privacy issues become major concerns because third-party cloud service providers are not always trusty. Although data contents can be protected by encryption, the access patterns that contain important information are still exposed to clouds or malicious attackers. In this paper, we apply the ORAM algorithm to enable privacy-preserving access to big data that are deployed in distributed file systems built upon hundreds or thousands of servers in a single or multiple geo-distribu ted cloud sites. Since the ORAM algorithm would lead to serious access load unbalance among storage servers, we study a data placement problem to achieve a load balanced storage system with improved availability and responsiveness. Due to the NP-hardness of this problem, we propose a low-complexity algorithm that can deal with large-scale problem size with respect to big data. Extensive simulations are conducted to show that our proposed algorithm finds results close to the optimal solution, and significantly outperforms a random data placement algorithm.

With the explosive growth in data, using data mining techniques to mine association rules, and then to find valuable information hidden in big data has become increasingly important. Various existing data mmmg
techniques often through mining frequent itemsets to derive association rules and access to relevant knowledge, but with the rapid arrival of the era of big data, Traditional data mining algorithms have been unable to meet large data's analysis needs. In view of this, this paper proposes an adaptation to the big data mining parallel algorithms-MRPrePost. MRPrePost is a parallel algorithm based on Hadoop platform, which improves PrePost by way of adding a prefix pattern, and on this basis into the parallel design ideas, making MRPrePost algorithm can adapt to mining large data's association rnles. Experiments show that MRPrePost algorithm is more superior than PrePost and PFP in terms of performance, and the stability and scalability of algorithms are better.

Cloud computing opens a new era in IT as it can provide various elastic and scalable IT services in a pay-as-you-go fashion, where its users can reduce the huge capital investments in their own IT infrastructure. In this philosophy, users of cloud storage services no longer physically maintain direct control over their data, which makes data security one of the major concerns of using cloud. Existing research work already allows data integrity to be verified without possession of the actual data file. When the verification is done by a trusted third party, this verification process is also called data auditing, and this third party is called an auditor. However, such schemes in existence suffer from several common drawbacks. First, a necessary authorization/authentication process is missing between the auditor and cloud service provider, i.e., anyone can challenge the cloud service provider for a proof of integrity of certain file, which potentially puts the quality of the so-called ‘auditing-as-aservice’ at risk; Second, although some of the recent work based on BLS signature can already support fully dynamic data updates over fixed-size data blocks, they only support updates with fixed-sized blocks as basic unit, which we call coarsegrained updates. As a result, every small update will cause re-computation and updating of the authenticator for an entire file block, which in turn causes higher storage and communication overheads. In this paper, we provide a formal analysis for possible types of fine-grained data updates and propose a scheme that can fully support authorized auditing and fine-grained update requests. Based on our scheme, we also propose an enhancement that can dramatically reduce communication overheads for verifying small updates. Theoretical analysis and experimental results demonstrate that our scheme can offer not only enhanced security and flexibility, but also significantly lower overhead for big data applications with a large number of frequent small updates, such as applications in social media and business transactions.

The explosive growth of demands on big data processing imposes a heavy burden on computation, storage, and communication in data centers, which hence incurs considerable operational expenditure to data center providers. Therefore, cost minimization has become an emergent issue for the upcoming big data era. Different from conventional cloud services, one of the main features of big data services is the tight coupling between data and computation as computation tasks can be conducted only when the corresponding data is available. As a result, three factors, i.e., task assignment, data placement and data movement, deeply influence the operational expenditure of data centers. In this paper, we are motivated to study the cost minimization problem via a joint
optimization of these three factors for big data services in geo-distributed data centers. To describe the task completion time with the consideration of both data transmission and computation, we propose a two-dimensional Markov chain and derive the average task completion time in closed-form. Furthermore, we model the problem as a mixed-integer non-linear programming (MINLP) and propose an efficient solution to linearize it. The high efficiency of our proposal is validated by extensive simulation based studies.