Browsing by Author "Roy, Arindam"

Now showing 1 - 9 of 9

Can Ensemble Machine Learning Methods Predict Stock Returns for Indian Banks Using Technical Indicators?
(MDPI, 2022) Roy, Arindam
This paper develops ensemble machine learning models (XGBoost, Gradient Boosting, and AdaBoost in addition to Random Forest) for predicting stock returns of Indian banks using technical indicators. These indicators are based on three broad categories of technical analysis: Price, Volume, and Turnover. Various error metrics like Mean Absolute Error (MAE), Mean Squared Error (MSE), Mean Absolute Percentage Error (MAPE), Root-Mean-Squared-Error (RMSE) have been used to check the performance of the models. Results show that the XGBoost algorithm performs best among the four ensemble models. The mean of absolute error and the root-mean-square -error vary around 3–5%. The feature importance plots generated by the models depict the importance of the variables in predicting the output. The proposed machine learning models help traders, investors, as well as portfolio managers, better predict the stock market trends and, in turn, the returns, particularly in banking stocks minimizing their sole dependency on macroeconomic factors. The techniques further assist the market participants in pre-empting any price-volume action across stocks irrespective of their size, liquidity, or past turnover. Finally, the techniques are incredibly robust and display a strong capability in predicting trend forecasts, particularly with any large deviations.
Enabling Workforce Optimization in Constrained Attribute-Based Access Control Systems
(IEEE, 2019) Roy, Arindam
Effective utilization of human capital is one of the key requirements for any successful business endeavor, with reorganization necessary if there are nonproductive employees or employees that are retiring. However, while reorganizing tasks for newer employees, it should be ensured that the employees have the requisite capabilities of handling the assigned tasks. Furthermore, security constraints forbid any arbitrary assignment of tasks to employees and also enforce major dependencies on other employees who have access to the same tasks. Since Attribute Based Access Control (ABAC) is poised to emerge as the de facto model for specifying access control policies in commercial information systems, we consider organizational policies and constraints to be modeled with ABAC. Given the increasing size and scale of organizations, both in terms of employees and resources that need to be managed, it is crucial that computational solutions are developed to automate the process of employee to task assignment. In this work, we define the Employee Replacement Problem (ERP) which answers the question of whether a given set of employees can be replaced by a smaller set of employees, while ensuring that the desired security constraints are not violated. We prove that the problem is NP-hard and use CNF-SAT to obtain a solution. An extensive experimental evaluation is carried out on diverse data sets to validate the efficiency of the proposed solution.
Impact of Multiple t-t SMER Constraints on Minimum User Requirement in RBAC
(Springer, 2014) Roy, Arindam
Separation of Duty (SoD) constraints are widely used to specify Role Based Access Control (RBAC) policies in commercial applications. It has been shown previously that efficient implementation of SoD policies in RBAC can be done using t-t Statically Mutually Exclusive Roles (SMER) constraints. In this paper, we present a method for finding the minimum number of users required under multiple t-t SMER constraints. The problem is shown to be NP-complete. We model the general problem using graphs, and present a two-step method for solving it. In the first step, a greedy algorithm is proposed that selects a graph which is likely to have the minimum chromatic number out of a set of graphs. The second step uses a known chromatic number finding algorithm for determining the chromatic number of the graph selected in the first step. Results for different values of the number of roles and the number of constraints as well as for different values of t have been reported.
Minimizing Organizational User Requirement while Meeting Security Constraints
(ACM Digital Library, 2015-09) Roy, Arindam
Large systems are complex and typically need automatic configuration to be managed effectively. In any organization, numerous tasks have to be carried out by employees. However, due to security needs, it is not feasible to directly assign any existing task to the first available employee. In order to meet many additional security requirements, constraints such as separation of duty, cardinality and binding have to be taken into consideration. Meeting these requirements imposes extra burden on organizations, which, however, is unavoidable in order to ensure security. While a trivial way of ensuring security is to assign each user to a single task, business organizations would typically like to minimize their costs and keep staffing requirements to a minimum. To meet these contradictory goals, we define the problem of Cardinality Constrained-Mutually Exclusive Task Minimum User Problem (CMUP), which aims to find the minimum users that can carry out a set of tasks while satisfying the given security constraints. We show that the CMUP problem is equivalent to a constrained version of the weak chromatic number problem in hypergraphs, which is NP-hard. We, therefore, propose a greedy solution. Our experimental evaluation shows that the proposed algorithm is both efficient and effective.
Minimum user requirement in Role Based Access Control with Separation of Duty constraints
(IEEE, 2013) Roy, Arindam
Constraints, specifically Separation of Duty (SoD) constraints, constitute an essential component for specifying Role Based Access Control (RBAC) policies. While it has been shown earlier that SoD constraints can be effectively represented using a set of t - t Statically Mutually Exclusive Roles (SMER) constraints, this paper presents a method for finding minimum number of users under multiple SMER constraints. We show that one way of solving the problem is to evaluate chromatic numbers for a set of graphs. However, since exhaustive search is computationally quite expensive, we present a genetic algorithm formulation of the problem. Each chromosome is a string of positive integers within a certain range and its length equals the number of t-t SMER constraints in the system. We report our results for different values of the number of roles and the number of constraints and also for different values of t.
On Optimal Employee Assignment in Constrained Role-Based Access Control Systems
(ACM Digital Library, 2016-12) Roy, Arindam
Since any organizational environment is typically resource constrained, especially in terms of human capital, organization managers would like to maximize the utilization of available human resources. However, tasks cannot simply be assigned to arbitrary employees since the employee needs to have the necessary capabilities for executing a task. Furthermore, security policies constrain the assignment of tasks to employees, especially given the other tasks assigned to the same employee. Since role-based access control (RBAC) is the most commonly used access control model for commercial information systems, we limit our attention to consider constraints in RBAC. In this article, we define the Employee Assignment Problem (EAP), which aims to identify an employee to role assignment such that it permits the maximal flexibility in assigning tasks to employees while ensuring that the required security constraints are met. We prove that finding an optimal solution is NP-complete and therefore provide a greedy solution. Experimental evaluation of the proposed approach shows that it is both efficient and effective.
Optimal Employee Recruitment in Organizations under Attribute-Based Access Control
(ACM Digital Library, 2021-01) Roy, Arindam
For any successful business endeavor, recruitment of a required number of appropriately qualified employees in proper positions is a key requirement. For effective utilization of human resources, reorganization of such workforce assignment is also a task of utmost importance. This includes situations when the under-performing employees have to be substituted with fresh applicants. Generally, the number of candidates applying for a position is large, and hence, the task of identifying an optimal subset becomes critical. Moreover, a human resource manager would also like to make use of the opportunity of retirement of employees to improve manpower utilization. However, the constraints enforced by the security policies prohibit any arbitrary assignment of tasks to employees. Further, the new employees should have the capabilities required to handle the assigned tasks. In this article, we formalize this problem as the Optimal Recruitment Problem (ORP), wherein the goal is to select the minimum number of fresh employees from a set of candidates to fill the vacant positions created by the outgoing employees, while ensuring satisfiability of the specified security conditions. The model used for specification of authorization policies and constraints is Attribute-Based Access Control (ABAC), since it is considered to be the de facto next-generation framework for handling organizational security policies. We show that the ORP problem is NP-hard and propose a greedy heuristic for solving it. Extensive experimental evaluation shows both the effectiveness and efficiency of the proposed solution.
Role Mining in the Presence of Separation of Duty Constraints
(Springer, 2015-12) Roy, Arindam
In recent years, Role Based Access Control (RBAC) has emerged as the most popular access control mechanism, especially for commercial applications. In RBAC, permissions are assigned to roles, which are then assigned to users. The key to the effectiveness of RBAC is the underlying role set that is used. The process of identifying an appropriate set of roles that optimally meets the organizational requirements is called role mining. One of the most useful constraints that can be expressed in RBAC is Separation of Duty (SoD). SoD constraints allow organizations to put a restriction on the minimum number of users required to complete a critical task. However, existing role mining algorithms do not handle SoD constraints and cannot be easily extended to incorporate SoD constraints. In this paper, we consider the problem of role mining when SoD constraints are present. We develop three alternative approaches that can be applied either during or after role mining. We evaluate the performance of all three approaches on several real world data sets and demonstrate their effectiveness.
A web enabled system designed for managing Hammersmith Infant Neurological Examination
(IEEE, 2013) Roy, Arindam
Information Systems plays a very important role in modernization and improvement of work flow management of health care units. This paper proposes a secure web enabled system based on a multi-tier architecture to conduct Hammersmith Infant Neurological Examination. The system also manages the follow-up patients efficiently. It contains features, which make it able to be used as training kit for medical professionals. A HINE performance monitoring interface is also provided with the system, which gives a brief idea to the doctors regarding the health of the infants