V. N. Sastry

Recurrent neural network and a hybrid model for prediction of stock returns

A novel hybrid model is proposed for prediction of stocks returns.The proposed model is hybrid of two linear models and a non-linear model.An optimization model is introduced which generates weights for proposed model.Proposed model is able to capture non-linear patterns of stock data very well. In this paper, we propose a robust and novel hybrid model for prediction of stock returns. The proposed model is constituted of two linear models: autoregressive moving average model, exponential smoothing model and a non-linear model: recurrent neural network. Training data for recurrent neural network is generated by a new regression model. Recurrent neural network produces satisfactory predictions as compared to linear models. With the goal to further improve the accuracy of predictions, the proposed hybrid prediction model merges predictions obtained from these three prediction based models. An optimization model is introduced which generates optimal weights for proposed model; the model is solved using genetic algorithms. The results confirm about the accuracy of the prediction performance of recurrent neural network. As expected, an outstanding prediction performance has been obtained from proposed hybrid prediction model as it outperforms recurrent neural network. The proposed model is certainly expected to be a promising approach in the field of prediction based models where data is non-linear, whose patterns are difficult to be captured by traditional models.

New polynomial time algorithms to compute a set of Pareto optimal paths for multi-objective shortest path problems

Multi-objective shortest path problem (MOSPP) is an active area of research because of its application in a large number of systems such as transportation systems, communication systems, power transmission systems, pipeline distribution systems of water, gas, blood and drainage, neural decision systems, production planning and project planning. In these networks it becomes necessary to find the best path from one node to a specified or all other nodes. The computational complexity of the existing algorithms in the literature to compute all Pareto minimum paths from a specified source node to all other nodes in an MOSPP is of exponential order in the worst case. Instead of generating all the values of the Pareto minimum paths in exponential time, we propose an algorithm to find a set of values of the Pareto minimum paths in polynomial time, which is very significant in many contexts. If an MOSPP of a network is having negative cycle, all the existing algorithm only indicate the existence of the negative cycle, that too after exponential number of operations. However, applying the proposed algorithm, we can find a set of Pareto minimum paths of any MOSPP of a network even if it contains negative cycles. The proposed algorithm is illustrated with examples.

New Algorithms For Multi Objective Shortest Path Problem

In recent years there has been an increase in research activity on multi-objective network optimization problems. Network optimization models can be obtained from a large number of application domains such as transportation systems, communication systems, pipeline distribution systems, fluid flow systems and neural decision systems. The primary aim of these network models is to optimize the performance with respect to pre defined objectives. Multiple objectives such as optimization of cost, time, distance, delay, risk, reliability, quality of service and environment impact etc. may arise in such problems. Many real life applications, dealing with above networks, require the computation of best or shortest paths from one node to another, called Shortest Path Problem (SPP). In this paper, three new algorithms for Multiple Objective Shortest Path Problem (MOSPP) and an algorithm to detect negative cycle in a network are proposed. MOSPP in a cyclic and acyclic network having weights either positive or negative or both can be solved using the proposed algorithms. Maximum number of Pareto optimal paths of a MOSPP in a network, is very much useful in finding the maximum number of iterations and the complexity of a particular algorithm. We prove here, the maximum number of Pareto optimal paths of any MOSPP in a completely connected network, in the worst case, is 1+(n−2)+(n−2)(n−3)+…+(n−2)!+(n−2)! and it lies between 2[(n−2)!] and 3[(n−2)!]. The computational complexities of the proposed algorithms have been analyzed. All proposed algorithms are illustrated with examples of cyclic and acyclic network.

STAMBA: Security Testing for Android Mobile Banking Apps

Mobile banking activity plays a major role for M-Commerce (Mobile-Commerce) applications in our daily life. With the increasing usage on mobile phones, vulnerabilities against these devices raised exponentially. The privacy and security of confidential financial data is one of the major issues in mobile devices. Android is the most popular operating system, not only to users but also for companies and vendors or (developers in android) of all kinds. Of course, because of this reason, it’s also become quite popular to malicious adversaries. For this, mobile security and risk assessment specialists and security engineers are in high demand. In this paper, we propose STAMBA (Security Testing for Android Mobile Banking Apps) and demonstrate tools at different levels. These supported tools are used to find threats at a mobile application code level, communication or network level, and at a device level. We give a detailed discussion about vulnerabilities that help design for further app development and a detailed automated security testing for mobile banking applications.

Modified algorithm to compute Pareto-optimal vectors

Finding Pareto-minimum vectors among r given vectors, each of dimension m, is a fundamental problem in multiobjective optimization problems or multiple-criteria decision-making problems. Corley and Moon (Ref. 1) have given an algorithm for finding all the Pareto-minimum paths of a multiobjective network optimization problem from the initial node to any other node. It uses another algorithm by Corley and Moon, which actually computes the Pareto-minimum vectors. We observed that the latter algorithm is incorrect. In this note, we correct the algorithm for computing Pareto-minimum vectors and present a modified algorithm.

Secure mobile payment framework based on UICC with formal verification

In this paper, we propose a secure mobile payments framework based on universal integrated circuit card UICC by defining: a a procedure of personalising UICC by the client; b a procedure of provisioning and personalisation mutual authentication and key agreement protocol of mobile payments application which is on UICC by the bank; and c a mobile payment protocol between the personalised mobile payment application on UICC and the bank server. Our provisioning and personalisation procedure is compared with recent works and found to be better in terms of generating clients credentials, implementation of WPKI in UICC, personalisation of mobile payment application by the bank and end to end security. Our mobile payment protocol originating from mobile payment application to the bank is also compared with recent works and found to be better in terms of confidentiality, authentication, integrity and non-repudiation, preventing double spending, over spending and money laundering, and withstands replay, man in the middle MITM and impersonation attacks. Proposed protocols are experimentally verified using BAN logic and scyther tool.

A review of adaptive approaches to MapReduce scheduling in heterogeneous environments.

MapReduce is currently a significant model for distributed processing of large-scale data intensive applications. MapReduce default scheduler is limited by the assumption that nodes of the cluster are homogeneous and that tasks progress linearly. This model of MapReduce scheduler is used to decide speculatively re-execution of straggler tasks. The assumption of homogeneity does not always hold in practice. MapReduce does not fundamentally consider heterogeneity of nodes in computer clusters. It is evident that total job execution time is extended by the straggler tasks in heterogeneous environments. Adaptation to Heterogeneous environment depends on computation and communication, architectures, memory and power. In this paper, first we explain about existing scheduling algorithms and their respective characteristics. Then we review some of the approaches of scheduling algorithms like LATE, SAMR and ESAMR, which have been aimed specifically to make the performance of MapReduce adaptive in heterogeneous environments. Additionally, we have also introduced a novel approach for scheduling processes for MapReduce scheduling in heterogeneous environments that is adaptive and thus learns from past execution performances.

Dynamic delegation approach for access control in grids

Access control is a mechanism to secure resources from unauthorized use. Securing a grid resource imposes a distinctive set of challenges to access control policies and mechanisms. As the grid resources are distributed in space and time, direct authorization methods are not always sufficient. Delegation is a way of indirect authorization whereby an active entity in a grid system can access a resource or act on behalf of another active entity. Access to resources in grids can be broadly categorized as full, nil or partial implying a degree of fuzziness. Conventional security models are rigid and the tasks that need to be performed by an active grid entity, require a more flexible form of access. A fuzzy logic based scheme provides more flexibility to resource access and access control in grids. The delegated rights have to be granted in a dynamic environment where the entities of a grid do not have prior information about each other. We propose a new scheme for this, which uses a two-stage fuzzy inference process based on trust relationships among the grid entities

Grid Security Through Delegation of Roles

Grid delegation is the process by which a user endows a grid-enabled program with the authority to the requisition and use of computational resources on behalf of the user. The large and geographically distributed, dynamic, heterogeneous and scalable grid environment poses unique delegation requirements. Presently the standard mechanisms to guide grid delegation are inadequate in their support for delegation. Since Globus tool kit-supported credential delegation is inadequate in a dynamic grid environment, a new conceptual model is required to effectively formulate grid delegation requirements. Viewing grid delegation as an authorization issue, we present a flexible framework called role-based grid delegation model (RB-GDM) for delegating access rights through roles in grids. It supports inter-domain and intra-domain role-based delegation in grids

A secure end-to-end SMS-based mobile banking protocol

Summary Short message service (SMS) provides a wide channel of communication for banking in mobile commerce and mobile payment. The transmission of SMS is not secure in the network using global system for mobile communications or general packet radio service. Security threats in SMS restricted the use of SMS in mobile banking within certain limits. This paper proposed a model to address the security of SMS using elliptic curve cryptography. The proposed model provides end-to-end SMS communication between the customer and the bank through the mobile application. The main objective of the proposed model is to design and develop a security framework for SMS banking. Further, the protocol is verified for its correctness and security properties because most of the protocols are not having the facility to be verified by using the formal methods. Our proposed framework is experimentally validated by formal methods using model checking tool called automated validation of internet security protocols and Scyther tools. Security analysis shows that the proposed mechanism works better compared to existing SMS payment protocols for real-world applications.