National Conference on Advancements in Information Technology

Abstract: Security worry for the transportation business in numerous nations. In the 11 years Rail mischances represent a critical from 2001 to 2012, the U.S. had more than 40000 rail mischances that cost more than $45 million .While a large portion of the mishaps amid this period had next to no cost, around 5200 had harms in abundance of $141500. To better comprehend the supporters of these extraordinary mishaps, the FederalRailroad Administration has required the railways required in mischances to submit reports that contain both ?xed ?eld sections and stories that portray the qualities of the mischance. While various reviews have taken a gander at the ?xed ?elds, none have done a broad examination of the accounts. This paper portrays the utilization of content mining with a mix of methods to naturally find mischance attributes that can advise a superior comprehension of the supporters of the mishaps. The review assesses the ef?cacy of content mining of mishap accounts by evaluating prescient execution for the expenses of extraordinary mischances. The outcomes how that prescient exactness for mischance costs signi?cantly enhances using highlights found by content mining and prescient precision additionally enhances using present day group techniques. Imperatively, this review likewise appears through case cases how the ?ndings from content mining of the stories can enhance comprehension of the supporters of rail mishaps in ways unrealistic through just ?xed ?eld examination of the mischance reports. Keywords: Security, extraordinary mischances, prescient execution, mischance reports.

Abstract: As an ever increasing number of uses deliver gushing information, clustering data streams has turned into an essential strategy for information and learning designing. A typical approach is to abridge the data streams progressively with an online procedure into an extensive number of alleged smaller scale bunches (micro-clusters). Micro-clusters are representatives for set of similar data points and are created using a single pass over the data. A conventional clustering algorithm is used in a second offline step to re-cluster the micro-clusters into final clusters sometimes referred to as macro-clusters. This paper depicts Novel Selection, is applied to the medical datasets which has many attributes. In the online stage for the selected disease name in the dataset micro-clusters are formed whereas in offline stage the doctor name is chosen for the selected disease, so that the macro-clusters are formed. This is done by the concept of shared density between the clusters i.e, which are similar to selected attributes, so the large number of smaller clusters will be created. Graph is plotted for both the clusters and also for the accuracy. The clustering quality will be increased by using shared density concept. Keywords: Data mining, data stream clustering, density-based clustering.

Abstract: As an ever increasing number of uses deliver gushing information, clustering data streams has turned into an essential strategy for information and learning designing. A typical approach is to abridge the data streams progressively with an online procedure into an extensive number of alleged smaller scale bunches (micro-clusters). Micro-clusters are representatives for set of similar data points and are created using a single pass over the data. A conventional clustering algorithm is used in a second offline step to re-cluster the micro-clusters into final clusters sometimes referred to as macro-clusters. This paper depicts Novel Selection, is applied to the medical datasets which has many attributes. In the online stage for the selected disease name in the dataset micro-clusters are formed whereas in offline stage the doctor name is chosen for the selected disease, so that the macro-clusters are formed. This is done by the concept of shared density between the clusters i.e, which are similar to selected attributes, so the large number of smaller clusters will be created. Graph is plotted for both the clusters and also for the accuracy. The clustering quality will be increased by using shared density concept. Keywords: Data mining, data stream clustering, density-based clustering.

Abstract: In the era of big data, many users and companies start to move their data to cloud storage to simplify data man-agement and reduce data maintenance cost. HoIver, security and privacy issues become major concerns because third-partycloud 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, I 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-distributed cloud sites. Since the ORAM algorithm would lead to serious access load unbalance among storage servers, I 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, I 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 my proposed algorithm finds results close to the optimal solution, and significantly outperforms a random data placement algorithm. Keywords: ORAM algorithm, big data, NP-hardness, random data placement algorithm.

Abstract: The innovation in cloud computing has encouraged the data owners to outsource their data managing system from local sites to profitable public cloud for excessive flexibility and profitable savings. But people can like full benefit of cloud computing, if we are able to report very real secrecy and security concerns that come with loading sensitive personal information. Allowing an encrypted cloud data search facility is of great significance. In view of the huge number of data users, documents in the cloud, it is important for the search facility to agree multi keywords query and arrange for result comparison ranking to meet the actual need of data recovery search and not regularly distinguish the search results. Related mechanisms on searchable encryption emphasis on single keyword search or Boolean keyword search, and often sort the search outcomes. In this system, we explain and solve the interesting problem of privacy preserving multi keywords ranked search over encrypted cloud data, and create a set of strict privacy necessities for such a safe cloud data application system to be effected in real. We first offer a basic idea for the multi keyword ranked. Keywords: ingle keyword search, Boolean keyword search, encrypted cloud data, multi keyword ranked.

Abstract: Blood is one of the important component of the body. It consists of RBC, WBC and Platelets. Detection and counting of white blood cells (WBC) in blood samples provides valuable information to hematologists, to identify various types of hematic pathologies such as AIDS and blood cancer (Leukemia). But performing this task manually prone to error and time consuming. An automatic detection and classification of WBC images can enhance the accuracy and speed up the detection of WBCs. In this paper, we propose an efficient framework for localization of WBCs within microscopic blood smear images using a multi-class ensemble classification mechanism. In the proposed framework, the nuclei are first segmented, followed by extraction of features such as texture, statistical, and wavelet features. Finally, the detected WBCs are classified into five classes including basophil, eosinophil, neutrophil, lymphocyte, and monocyte. The proposed method improves the segmentation performance when compared to other state-of-the-art segmentation methods. Keywords: Hematology; Image Segmentation; Image Classification;multi-class ensemble.

Abstract: In distributed computing infrastructure, even if specialized high performance server hardware might provide better performance per watt ratios, a classical server will consume more power when it is idle, it will also need more supporting infrastructure such as space, air conditioning ,and thus imposes a higher total cost of ownership(TCO) compared to mobile devices and to setup a new such infrastructure will indeed require a huge amount in terms of both economically and human resources. Every night, many smartphones are plugged into a power source for recharging the battery. Given the increasing computing capabilities of smartphones, these idle phones constitute a size able computing infrastructure. Therefore, for an enterprise which supplies its employees with smartphones, it is arguable that a computing infrastructure that leverages idle smartphones being charged overnight is an energy - efficient and cost-effective alternative to running certain tasks on traditional servers. Keywords: Smartphone, Infrastructure, Computing.

Abstract: The human visual system is one of the wonders of the world. The difficulty of visual pattern recognition becomes apparent if you attempt to write a computer program to recognize digits. One of the approaches to computers behaving and computing like humans involve neural networks. The neural network automatically infers rules for recognizing handwritten digits. This approach reduces human intervention in many commercial places like banks to process cheques and by post offices to recognize addresses. The neural network attempts to determine if the input data matches a pattern that the neural network has memorized. The concepts of weights and biases are conventionally used to enhance the performance of the Neural networks over a large set of test images and classify the digit into a class label. The data is stored in a very simple file format designed for storing vectors and multidimensional matrices. Tensor Flow is used to train an elaborate model that achieves state-of-the-art performance along with Softmax Regression. The current program can recognize digits with an accuracy over 96 percent, without human intervention, classifying 9,979 of 10,000 images correctly. The performance is close to human-equivalent, and is arguably better, since quite a few of the MNIST images are difficult even for humans to recognize with confidence. Keywords: Neural network; TensorFlow; Softwax Regression; Pattern recognition.

Abstract: Compromisation of online social network is a threat for many of us who are a part in OSN. Where many spammers establish and achieve our trust of friends and success in sending malicious spams and try to hack our account. In this paper, our goal is to analysis the social behaviour of such attackers and user, by usage of OSN services. We propose a set of social behaviouralfeatures that can effectively characterize the user social activitieson OSNs. We validate the efficacy of these behavioural features by collecting and analysing real user clickstreams to anOSN website. Based on our measurement study, we devise individual user's social behavioural profile by combining its respectivebehavioural feature metrics. A social behavioural profile accuratelyreflects a user's OSN activity patterns. While an authentic ownerconforms to its account's social behavioural profile involuntarily,it is hard and costly for impostors to feign. We evaluate thecapability of the social behavioural profiles in distinguishingdifferent OSN users, and our experimental results show thesocialbehavioural profiles can accurately differentiate individualOSN users and detect compromised accounts. Keywords: Online social behavior, privacy, data analysis,compromised accounts detection.

Abstract: As a side effect of increasingly popular social media, cyberbullying has emerged as a serious problem af?icting children, adolescents and young adults. Machine learning techniques make automatic detection of bullying messages in social media possible, and this could help to construct a healthy and safe social media environment. In this meaningful research area, one critical issue is robust and discriminative numerical representation learning of text messages. In this paper, a new representation learning method is introduced to tackle this problem. The method named Semantic-Enhanced Marginalized Denoising Auto-Encoder (smSDA) is developed via semantic extension of the popular deep learning model stacked denoisingautoencoder. The semantic extension consists of semantic dropout noise and sparsity constraints, where the semantic dropout noise is designed based on domain knowledge and the word embedding technique. The proposed method is able to exploit the hidden feature structure of bullying information and learn a robust and discriminative representation of text. Keywords: Cyberbullying Detection, Text Mining, Representation Learning, Stacked DenoisingAutoencoders, Word Embedding.

Abstract: The main reason for new forms of crime labeled cyber crime is the speedy growth of the internet and computer technology over the past few ages.Cyber is the imaginary space, which is created when the electronic devices communicate, similar to network of computers, Cyber crime refers to anything done in the cyber space with a criminal intent. These could be either the criminal activities in the traditional sense or could be activities which are newly evolved with the growth of the technology. Cyber crime includes acts such as hacking, uploading obscene content on the Internet, sending obscene e-mails(spamming) and hacking into a person's e-banking account to withdraw money(phishing) The concept of cyber crime is not very different from the concept of conservative crime since both include conduction of unauthorised action or omission, which breaks the rules of law. Cybercrime has been one of the common practices made by the computer experts. this paper its mentioned about some of the impacts of the cybercrime. This paper gives detailed information regarding cybercrime, its types, modes of cybercrime and security measures including prevention to deal effectively with cybercrime. Keywords: e-mails(spamming), ithdraw money(phishing), cybercrime.

Abstract: This Decade, the boundaries between e-commerce and social networking have become increasingly blurred. Lots of e-commerce web Application support the process of social login where users can sign on the websites using their social network username and password authentication such as their Twitter or Facebook accounts. Social Network users can also post their newly purchased products on microblogs with links to the e-commerce product web pages. In this paper, we propose a novel solution for cross-site cold-start product recommendation. We aim to recommend e-commerce product from e-commerce websites to users at social networking websites in "cold-start" situations. Cold-start situation is a problem which has rarely been explored before. A major challenge is how to leverage knowledge extracted from social networking sites for cross-site cold-start product recommendation. We propose to use the linked users across social networking sites and e-commerce websites as a bridge to map users' social networking features to another feature representation for product recommendation. In specific, we propose learning both users' and products' feature representations from data collected from e-commerce websites using recurrent neural networks and then apply a modified gradient boosting trees method to transform users' social networking features into user embeddings. We then develop a feature-based matrix factorization approach which can leverage the learnt user embeddings for cold-start product recommendation. Experimental results on a large dataset constructed from the microblogging service FACEBOOK and the largest e-commerce website AMAZON have shown the effectiveness of our proposed framework. Keywords: Cold start, Product Recommendation, E-commerce, Micro-blogs, Product Demography, Data mining, Information Search.

Abstract: The paper deals with the development of a cloud integration service that involves building multiple integration components. An integration component is a particular, metamorphic piece of functionality that consistently represents a way to communicate with one organization and/or API. Sharing information and providing visibility into both frontend and backend systems using cloud integration increases productivity and simplifies business process. With third-party cloud service integration, organizations can focus more on driving new business and less on the hassles of trying to make data available. Moreover, synchronizing data to provide updated information whenever a change is made in either of the cloud services can be automated, with valuable data from cloud service systems available to the right people at the right time, all through a single interface. The system provides generic interface between components. Keywords: cloud integration; iPaaS; OAuth2.0; RESTful APIs; Accounting; e-Commerce.

Abstract: Our project is a Java implementation of Chaos based algorithm for fingerprint encryption. Biometric traits are unique to each person and wherever he goes, it goes with him. It is a very effective identification system. A biometric trait, such as fingerprint, palm-print, iris-scan, face-scan, etc., is taken as marks to identify a person, as does our human brain. We have taken up fingerprint as our biometric trait, to experiment with, in our project. Fingerprint authentication is an efficient system, as opposed to password-based authentication, where the password can be lost or forgotten. Nevertheless, the security of the user's data and his biometric trait are a concern as they are unique to each person and can lead to identity theft if it's details get into the wrong hands. Hence, came forth the technique of encrypting sensitive data, so that, a random person who comes across this data will not be able to tamper with it, because, this data will not make any sense to him. When the authorized user needs it, it will be decrypted. A matching process will be implemented to match the provided user's fingerprint with that of the encrypted stored data, to cross check if the two are the same and grant access accordingly. There are several approaches in the recent years that have provided biometric revocability feature, but lack robustness and security. Chaos-based systems, on the other hand, have a good number of properties that are defined in the ideal features of secure cryptography, i.e., extreme sensibility to initial conditions withConfusion and, ergodicity with diffusion. We implement chaos based algorithm, using Java programing language, as a working software, where the system interacts with the user and allows him to register or sign-in to his account. The program is also connected to a securedatabase where the encrypted user data is stored. The scanning of image is done with the help of a fingerprint scanner. We use SMTP (JavaMail) to send an e-mail to the user if there has been any malicious activity detected on his/her account.This system is secure, effective at low cost and can be implemented on real, secure access control systems. Keywords: Chaos-based systems, secure access control systems, SMTP.

Abstract: Due to advent of new technologies, the amount of data produced by mankind has already reached a zetta-byte level. New devices, businesses and social networking sites are a major source for the production of such large amount of data. This data is really huge and collectively called by a well known term "Big data". Due to such huge amount of data being available it becomes very difficult to perform effective analysis using the currently available traditional techniques. From the literature survey it is found that there are a total of 39 tools available for analysis and processing of big data. Survey reveals that but the most influential and established tool for analyzing big data is Apache Hadoop which is an open source framework written in java that allows parallel processing across clusters of computers using basic programming techniques. This paper introduces apache Hadoop, its framework, installation and how it uses map reduce and cluster programming to capture, analyse and process big data." Keywords: big data, Hadoop, mapreduce, HDFS, Clustered.

Abstract: When we think of secure transmission of information, the term Cryptography comes to our mind. By using various encryption algorithm, decryption algorithm, key generation algorithm and key matching algorithm cryptography ensures secure transaction of information between the sender and the receiver without the intrusion of an attacker. In this paper primarily cryptography is merged with fingerprint recognition technology which is one of the main forms of Biometrics. Here an inherent biometric characteristic like fingerprints are used to generate the key. So it is more secure compared to symmetric encryption where a lot of care has to be taken in storing the key in a secure place. To put this concept into effect, we use the sender's fingerprint geometry for the generation of the key. On the receiver's side, database of the sender's fingerprint images will be present through which the decryption process takes place using fingerprint matching algorithms. This algorithm is applied on the binary conversion of the fingerprint images and this whole method is applicable on the binary form. Keywords: Cryptography, Encryption, Decryption, Key generation, Biometrics, Fingerprint geometry.

Abstract: For almost every industry, the impact of automation and artificial intelligence (AI) has become a significant topic of discussion as we enter the origin of a long-awaited future. The conversation nowadays is on elimination of work through robotics and going beyond those efficiencies to deliver on self-healing through artificial intelligence (AI) and machine learning. Customer understanding is becoming more key, with the customer here being defined generally as any set of stakeholders across any set of processes that are being serviced. These topics are limited to the impact on a current set of Business Process Outsourcing (BPO) processes and not the larger troublesome impact of digitization to the core business itself, since that's a conversation that transcends BPO. This paper gives an overview of the application of artificial intelligence and learning concepts in many IT organizations and BPOs. Keywords: Intelligent being, Data prediction, Envisioning the future.

Abstract: Cloud computing provides a cheap and resourceful solution for sharing group resources among cloud users at a low maintenance. In this current world, outsourcing data in multi owner fashion from un trusted cloud is still a tricky issue due to some frequent changes in membership. In this project, we propose a safe and sound data sharing process for vibrant groups in the cloud. By the usage of dynamic techniques, any cloud user can secretly share data with others, but here cloud will secure in an efficient way. The main objective of this project is that the cloud users can share their files such as pdf, txt, doc in a secure way by using the encryption and decryption methods. In this scenario, different Data owner will be in the cloud, those data owners can send the data to their group members in a secured way . Moreover, they will monitor their particular group activities. User will registered with their preferred group and share their files among that particular groups and if they like to share their file with all, there is an option like common sharing method. By using this option other group members can make use of it with proper verifications. Once user registered with valid data and with their preferred group selection. Data owner need to distribute group key to all their registered group members in a secure and authentic manner. Keywords: Cloud computing, group members, authentic manner.