C V P R PRASAD
@mallareddyecw.com
Professor and CSE
MALLA REDDY ENGINEERING COLLEGE FOR WOMEN
RESEARCH INTERESTS
Knowledge Engineering, Data Mining, Software Engineering
Scopus Publications
Scopus Publications
CH.M.H. Saibaba, Saiyed Faiayaz Waris, S.Hrushikesava Raju, VSRK Sarma, Vijaya Chandra Jadala, and Chitturi Prasad
IEEE
The basic plan of our work is to design and implement Artificial Intelligence based Voice Recognition System Software and to create a bundle that consists integration of several tasks, where we can process and execute using Clients Individual voice command. An Artificial Intelligent based Virtual or Personal Assistant were indicated as IVA or IPA could be a package of Intelligent mechanism which execute different tasks and services based on the queries and command for operating intelligent based system. It is identified as square measure at some areas of disciplines utilized for voice detection, where voice recognition is identified as communication channel between persons will be notified by machine. where machines can identify voice of a person wherever the voice is only approach of double communication, and a lot of typically, it permits to free-up each hands and vision doubtless for performing an additional action in comparison or facilitates conjointly inactivated individuals. In this paper we concentrated on the different artificial intelligent technologies that support the voice recognition and natural language processing. Generally, to try and do such quite tasks we would like a voice assistant to prefer and to get gadgets like Alexa. However identical practicality will be done by several the powerful packages like pywhatkit, Wikipedia, pyttsx3, pygame, speech recognition, OpenCV etc., we majorly focused on the practical implementation of voice recognition module based on the OpenCV. In modern technology speech recognition, object identification, digital image processing, language processing plays a major role in security and other factors, the most common mechanisms in artificial intelligence in association with machine and deep learning is that the mechanisms for identifying the natural language of a personalities. Latest terminology during present issue will cause artificial intelligence-based human-robot interaction that is based on multi-disciplinary technologies such as natural language processing, deep and machine learning. We even concentrated on the different tools that support voice recognition and the software companies such as google, Microsoft and amazon. Where new trends and methodologies are implemented at rapid growth. The major factors in this module are natural language processing, speech recognition and the voice management. One is every of such tools is voice assistant, which can be integrated into several alternative intelligent systems. Voice may be a ton of economical than writing on a keyboard. Here we tend to square measure desegregation varied options and place along as one practicable file. Hence, we will create several things among less quantity of your time with the simplest performance. In this research we verified different models, mechanisms and procedures adopted by various researchers in this field to recognize the voice. The basic set up of our work is to integrate as many tasks as achievable and build it execute by our voice command. The design and implementation of the OpenCV approach is given along with the experimental analysis and finally the result analysis is given in the graphical form.
Chitturi Prasad, Gummuluri Udaya Chandrika, Vani Venkata Sai Sindhu Garapati, Ganaga Rama Koteswara Rao, Vadavalli Harshitha Manaswini, and Pinnamaneni Meghana
IEEE
Immutability is a term used a lot in blockchain and cryptocurrency development when attemptingto analyze the economic interpretation of a network. Designs like the 21 million Bitcoin supply in BTC or incidents like the DAO fork bailout that split the Ethereum community into ETH and ETC are major examples of immutability values determining the ethos of a chain. This research sets out tocategorize different features of blockchain immutability and analyze them in three different networks: Bitcoin (BTC), Ethereum (ETH), and Ethereum Classic (ETC). Further in the paper, the Finney Ratio is introduced to help understand how blockchain immutability can be measured over time. The paper then visualizes overall blockchain immutability over time using radar charts as an immutability map. The Szabo Score is then introduced as a tool to score overall mutability over time for each blockchain.
Srilatha Yelamati, Dammalapati Ravi Kiran, V.V.S. Sasank, and Chitturi Prasad
IEEE
The implemented model constrains the beta coefficients for price and budget. Both are restricted to positive values only, which makes the model much more realistic. Usually one would assume the price coefficients to be negative only, however in the BLP-type specification there is a reward for not using up the entire budget. Hence, ${\\beta _{price}}$ must be positive (Pachali et al. 2017b). In some cases, it makes sense to additionally impose ordinal constraints on the betas among brands. This would be beneficial if there were a clear and strict preference for individual specifications of the products in question (i.e. a higher quality ceteris paribus is always more desirable). For this particular beer data set, this cannot reasonably be justified, since there is no objective order of beers. The research question whether a merger between Heineken and Amstel would yield an incremental profit beyond the sum of the individual profits can now be answered by changing the ownership matrix so that those three beers belong to the same owner and thus not price compete anymore. This market is now a duopoly with two players (Heineken together with Amstel vs. Estrella) and referred to as the merge competition scenario. Table 1 displays the weighted average price and the producer surplus before and after the merger and quantifies the absolute and relative difference between those two scenarios. While ${\\bar p}$ increases by 3.3%, the producer surplus even increased.
K P N V Satya Sree, Jayavarapu Karthik, Ch Niharika, P V V S Srinivas, N Ravinder, and Chitturi Prasad
IEEE
While some data have an explicit, numerical form, many other data, such as gender or nationality, do not typically use numbers and are referred to as categorical data. Thus, machine learning algorithms need a way of representing categorical information numerically in order to be able to analyze them. Our project specifically focuses on optimizing the conversion of categorical features to a numerical form in order to maximize the effectiveness of various machine learning models. From the methods utilized, it has been observed that wide and deep is the most effective model for datasets that contain high-cardinality features, as opposed to learn embedding and one-hot encoding.
Thulasi Bikku, Jayavarapu Karthik, Ganga Rama Koteswara Rao, K P N V Satya Sree, P V V S Srinivas, and Chitturi Prasad
IEEE
Deep convolutional neural networks were used to successfully segment several important neural tissue classes in MRI brain images, and approaches for integrating prior information into the networks to increase their performance on this task were investigated. Regrettably, only the first of them is addressed in this paper. To make the implementation of nonstandard architectures, which was expected to be required for the second goal, it was determined to provide a framework for defining and training networks by only using fundamental components. While this was an educational experience, the amount of progress accomplished was far less than if a conventional network package had been utilized instead. The requirement to deal with all of the lowest level aspects of network construction, from initialization schemes to adaptive learning rates and all the other components of the optimizer pipeline has left no time for utilizing this infrastructure to do something which has not been accomplished by the existing frameworks, and of course in all other respects it is far more limited than they are. It would in-stead be focused on a clear and detailed analysis of the full pipeline, which is required to build a network for solving the first problem. Despite several difficulties tracking down bugs in the optimizer, GPU memory allocation, and the last-minute accidental deletion of a large portion of the experimental results, the software implementation made available at: achieves DICE results of 0:8, which, while not class leading, would still place well in many benchmarks.
Rajesh Kumar Ojha, Sandeep Srivastava, Mohit Goyal, Lalan Kumar, Amit Kumar, and Chitturi Prasad
IEEE
K-Means is known both for its usefulness in finding clusters of related data as well as its fragility with respect to initialization choices. This paper introduces a 95% more effective and 50% more efficient initialization methods, that could eliminate the need for multiple executions of K-Means to find high quality clustering. To initialize the centroids, it selects a multiple, m, of K real data points, computes (mK)2 distances and keeps only the K maximum( minimum( distance ) ) points. A consequence of this technique enables O(lnK) binary search to find the optimal K on ’linearly’ separable clusters. The effectiveness claim applies both to separable and intertwined clusters although the efficiency is lost on intertwined clusters.
Rajesh Kumar Ojha, S Janardhana Rao, Pankaj Goel, Sandeep Srivastava, K. Hareesh Kumar, and Chitturi Prasad
IEEE
This paper presents a number of novel approaches to help facilitate Deep Reinforcement Learning (DRL) for Neural Networks based agents in the domain of General Video Game Playing (GVGP). Using common processing methods, the NN can retain a fixed predetermined input and output shape while having the flexibility to play a variety of games. Training can also be made more efficient via reward normalization. Our results show that these modifications negligibly impact the learning time and performance of the model when testedupon games in the GVGAI framework. Furthermore analysis of these approaches within the parameters of the GVGAI learning competition was also performed. This research work attempts to use the adaptations presented to train an agent thatwill compete in the CEC2019 GVGAI Learning Competition.
Neenavath Veeraiah, Osamah Ibrahim Khalaf, C. V. P. R. Prasad, Youseef Alotaibi, Abdulmajeed Alsufyani, Saleh Ahmed Alghamdi, and Nawal Alsufyani
Institute of Electrical and Electronics Engineers (IEEE)
Mobile ad hoc network (MANETs) is infrastructure-less, self-organizing, fast deployable wireless network, so they truly are exceptionally appropriate for purposes between special outside occasions, communications in locations without a radio infrastructure, crises, and natural disasters, along with military surgeries. Security could be the primary weak spot in manet on account of this flexibility of structures and always changing dynamic topology, that will be very exposed to your selection of strikes like eavesdropping, routing, and alteration of programs. MANET is affected with security issues, surpassing Quality of services (QoS). So, intrusion tracking which modulates your system to recognize some other violation weakness would be that the top approach to guarantee security for MANET. Detecting intrusions has a critical part in supplying protections and functions as beyond layer of defenses against access. Power collapse of the cellular node maybe not merely alter the node alone but its capacity to forwards packets which is based on total system life. This also caused the institution of the routing protocol to its stable optimal choice of this multi-path to increase the navigation MANETs. Provision of energy-efficient and secure routing is a challenge given the changing topology and restricted resources of this kind of network. To address the energy efficiency and security we suggest a trust-based secure energy efficient navigation in MANETs employing the hybrid algorithm, cat slap single-player algorithm (C-SSA), that selects the best jumps in advancing the routing. In the beginning, the fuzzy clustering is put on, and the cluster heads (CHs) are picked predicated maximum worth of indirect, direct, and recent trust. Predicated on trust threshold worth nodes additionally discovered. Even the CHs are participated from the multi hop routing, and the assortment of the best route relies upon the projected hybrid protocol, and that selects the best routes determined by the delay, throughput, along with connectivity within this course. The proposed method obtained a minimal energy of 0.11m joules, a negligible delay of 0.005 msec, a maximum throughput of 0.74 bps, a maximum packet delivery ratio of 0.99 %, and a maximum detection rate of 90%. The proposed method compared with the existing techniques in the presence and absent of the selective packet dropping attack.
K. B. V. Rama Narasimham, C. V. P. R. Prasad, J. Jyothirmai, and M. Raghava
Springer Singapore
Vector representations for language have been shown to be useful in a number of Natural Language Processing tasks. In this paper, we aim to investigate the effectiveness of word vector representations for the problem of Sentiment Analysis. In particular, we target three sub-tasks namely sentiment words extraction, polarity of sentiment words detection, and text sentiment prediction. We investigate the effectiveness of vector representations over different text data and evaluate the quality of domain-dependent vectors. Vector representations has been used to compute various vector-based features and conduct systematically experiments to demonstrate their effectiveness. Using simple vector based features can achieve better results for text sentiment analysis of APP.
In business intelligence, large number of data to be generated because of increasing data in business applications. Analysis and prediction of data is very aggressive concept to evaluate the results present in data based on decision making analysis. To provide effective analysis of data traditionally some of the machine learning related methods like Clustering, Classification, Neural network based approaches and association rule based approaches were used to explore and analysis of business data. Because of increasing depth analysis of data in business intelligence related applications then above static machine learning approaches were not satisfied to form association between different attributes in real time data sets. So that in this paper, we propose Advanced & Hybrid Machine Learning Approach (AHMLA) for effective data analysis of different associated attributes of high dimensional data. Our proposed approach increase customer service, report generations based on user awareness in business intelligence applications. An experimental result of proposed approach gives better high performance with respect to different parameters with respect to existing approaches
In data mining, mining and analysis of data from different transactional data sources is an aggressive concept to explore optimal relations between different item sets. In recent years number of algorithms/methods was proposed to mine associated rule based item sets from transactional databases. Mining optimized high utility (like profit) association rule based item sets from transactional databases is still a challenging task in item set extraction in terms of execution time. We propose High Utility based Association Pattern Growth (HUAPG) approach to explore high association utility item sets from transactional data sets based on user item sets. User related item sets to mine associated items using utility data structure (UP-tree) with respect to identification of item sets in proposed approach. Proposed approach performance with compared to hybrid and existing methods worked on synthetic related data sets. Experimental results of proposed approach not only filter candidate item sets and also reduce the run time when database contain high amount of data transactions.
Bhanu Prakash Battula, Debnath Bhattacharyya, C. V. P. R. Prasad, and Tai-hoon Kim
Springer International Publishing
Data Mining is a popular knowledge discovery technique. In data mining decision trees are of the simple and powerful decision making models. One of the limitations in decision trees is towards the data source which they tackle. If data sources which are given as input to decision tree are of imbalance nature then the efficiency of decision tree drops drastically, we propose a decision tree structure which mimics human learning by performing balance of data source to some extent. In this paper, we propose a novel method based on sampling strategy. Extensive experiments, using C4.5 decision tree as base classifier, show that the performance measures of our method is comparable to state-of-the-art methods.