Search

Trading strategies to maximize profits by tracking and responding to dynamic stock market variations is a complex task. This paper proposes to use a multilayer perceptron method (a part of artificial neural networks (ANNs)), that can be used to deploy deep reinforcement strategies to learn the process of predicting and analyzing the stock market products with the aim to maximize profit making. We trained a deep reinforcement agent using the four algorithms: proximal policy optimization (PPO), deep Q-learning (DQN), deep deterministic policy gradient (DDPG) method, and advantage actor critic (A2C). The proposed system, comprising these algorithms, is tested using real time stock data of two products: Dow Jones (DJIA-index), and Qualcomm (shares). The performance of the agent linked to the individual algorithms was evaluated, compared and analyzed using Sharpe ratio, Sortino ratio, Skew and Kurtosis, thus leading to the most effective algorithm being chosen. Based on the parameter values, the algorithm that maximizes profit making for the respective financial product was determined. We also extended the same approach to study and ascertain the predictive performance of the algorithms on trading under highly volatile scenario, such as the pandemic coronavirus disease 2019 (COVID-19).

Recommender systems are tools that support personalization in terms of supporting navigation, sharing, and discovery of information and help users to find their desired content over the large volume of information. Recently, new research area on context-aware recommendations has emerged to provide the capability of utilizing social contents and exploit related tags and rating information and personalize the search for desired content by considering user’s actual situation (contextual information). In this study, we propose an approach for clustering contextually similar information using unsupervised learning approach through K-Medoids clustering and demonstrate the extraction of latent preferences for recommending items under a given contextual cluster and study how such clusters of similar contextual information can be exploited to improve the prediction accuracy of a context-aware recommender systems. To evaluate the performance of our proposed recommendation strategy, the empirical analysis is conducted on the popular LDOS-CoMoDa dataset and we showed that our proposed approach outperforms state-of-the-art algorithms in terms of prediction accuracy of the computed recommendations.

Digitization of music and advancements in information technology for sharing information on World Wide Web paved way for its availability in enormous quantities anywhere any time. Rather than retrieving annotated music in response to query given in terms of Metadata such as name of the composer/singer, genre etc modern researchers are challenged towards content based music information retrieval systems (CBMIR). CBMIR systems differ while representing the main melody either as a note sequence or as an analog acoustic signal; the note sequence representation is explored in this research work. Based on the observation that repeating patterns of the note sequences representing the main melody capture the essence of the music object, this research work developed a framework to investigate the feasibility and effectiveness of pattern based melody matching approach to music information retrieval. Experimentation is conducted on a real world dataset of musical objects belonging to South Indian classical music and the performance of the framework is estimated in terms of Mean Reciprocal Ranking.

The process of rice breeding involves producing new rice varieties as a cross of parental rice varieties followed by rigorous testing and examination phase for purity, productivity and resistivity to regional climatic conditions. The selection of appropriate parental seeds to produce new rice variety with more or less predictable characteristics is highly desirable as it reduces expensive experimental evaluation efforts. Authors suggest the applicability of Conceptual Overlapping Clustering Algorithm [7] developed by them for conceptual overlapping clustering to aid proper selection of rice varieties and demonstrated the performance of this algorithm on a real world dataset.

Information security is necessary for any organization. Intrusion prevention is the basic level of security which requires user authentication. User can be authenticated to a machine by passwords. Traditional textual passwords are vulnerable to many attacks. Graphical passwords are introduced as alternatives to textual passwords to overcome these problems. This paper introduces native language passwords for authentication. Native language character set consists of characters with single or multiple strokes. User can select one (or more) character(s) for his password. The shape and strokes of the characters are used for authentication.

User authentication is the first phase of information security. Users should remember their passwords and recall them for authentication. Text based passwords is the traditional method for authentication. Short and simple passwords are memorable and usable but not secure. Random and lengthy passwords are secure but not memorable and usable. Graphical password schemes are introduced as alternatives to text based schemes. Few grid based authentication techniques are also proposed. The purpose of this paper is to introduce a tool to enhance the memorability and security of passwords which also provides usability. The most popular ancient Indian board game “Snakes and Ladders” is used as a tool for authentication.

Network security measures are required to protect data during their transmission. Sensitive information transmitted across a network requires confidentiality. Data confidentiality is the protection of transmitted data from passive attacks. Passive attacks are eaves dropping or monitoring the traffic. The goal of the intruder is to obtain information being transmitted. These attacks can be prevented by encryption. This paper proposes a simple grid based method for providing confidentiality to the data or the session key being transmitted. The sender and the receiver share a global secret key which is used for extracting the data from a grid.

Authentication is necessary in multi-user systems. User name and password are used to authenticate a user. Textual passwords are most common type used for authentication. Authentication schemes that use textual passwords are vulnerable to attacks password stealing, dictionary attack and shoulder surfing. Graphical passwords provide an alternative to graphical passwords but, simple schemes are vulnerable to shoulder surfing and hidden cameras. Native language passwords can be used for authentication and the user can remember it better than any other language. A shape based textual authentication is discussed in this paper and simplification of the native language passwords is proposed. Session passwords are generated for each login, making the authentication scheme more resistant to shoulder surfing and hidden camera attacks.

Periodic pattern is a pattern that repeats itself with a specific period in a given sequence. Patterns that occur frequently with strict periodicity in one or more subsequences separated by tolerable disturbance are called asynchronous periodic patterns. Longest Subsequence Identification (LSI) is the pioneering algorithm to mine asynchronous periodic patterns. For each asynchronous periodic pattern the algorithm detects the longest subsequence containing it. Simple Multiple Complex and Asynchronous periodic pattern miner (SMCA) is a four phase algorithm that detects all the subsequences containing asynchronous periodic patterns. One Event One Pattern (OEOP) algorithm uses a linked list structure to detect single event one patterns in a single scan of a sequence. OEOP can be used to replace the first phase of SMCA for data sets like data streams. When compared to SMCA, E-MAP can efficiently mine all patterns in a single step and single scan of the sequence in the presence of large primary memory.

In a multi-user system, user name and password serves to authenticate the user. Generally users select alphanumerical passwords or textual passwords in English. It is easy for the intruder to crack these passwords by eaves dropping, password stealing, dictionary attack and shoulder surfing. To overcome these vulnerabilities, graphical password schemes have been introduced. An intruder can easily break the simple graphical password authentication schemes by shoulder surfing and hidden cameras. In this paper a new shape based textual authentication scheme for native language passwords is proposed. User selects a character from his native language and the shape of this character becomes password criteria. The proposed authentication scheme is resistant to attacks like password stealing, eves dropping, shoulder surfing and hidden cameras because every time user enters a new password.