2017-18

Permanent URI for this collection

http://172.16.1.132:4000/handle/123456789/9408

Browse

Now showing 1 - 20 of 45

Analysing Public Response for Government Initiatives using Machine Learning
(2018-06-19T07:00:09Z) MIR KAMRAN, ABBAS; SAHANA, K; NEHAL, SAHU
Sentiment analysis or opinion mining is one of the major tasks of NLP (Natural Language Processing). Sentiment analysis has gain much attention in recent years. In this paper, we aim to tackle the problem of sentiment polarity categorization, which is one of the fundamental problems of sentiment analysis. A general process for sentiment polarity categorization is proposed with detailed process descriptions. Indian Government has launched major reforms like Swach Bharat, demonetization and twitter data sets about public opinion on these policies are extracted and sentiment analysis is done on it in this project work.
Assisting Crop Selection for Farmer
(2018-06-19T10:34:41Z) UMASHREE R, KADIWAL; MONISHA, V; SWATHI, M N
In Farming based countries like INDIA, it is required for farmers to predict the yield of crop to be planted well before planting. If yield can be predicted, farmers can calculate their profit margin and decide which crop to plant. The prediction is based on three macro factors rainfall, humidity and temperature. In this project, we propose a Neuro Fuzzy system to predict the yield of crop and a ARIMA model to predict the price of crop.
Automation Script Development for Powerchart
(2018-06-19T10:02:28Z) SAI RAMYA, S R; VISHWAS V, AITHAL; SAI SHREE, NAGIREDDY
The Project entitled “Automation Script Development for PowerChart” is automating the manual testing process currently in use. Automation is the use of strategies, tools and artifacts that augment or reduce the need of manual or human involvement or interaction in unskilled, repetitive or redundant tasks. Automation of plans are done using Organization’s tool called Touchstone 9.1. Touchstone 9.1 is an Organization’s internally-developed automation testing tool specifically used for UI and functional testing. Touchstone 9.1 used in creating automated regression tests to supplement or replace manual tests in order to complete testing faster and helps in insuring quality. The application PowerChart is a family of system solutions for a wide assortment of health care providers. As an electronic medical record system, PowerChart supports enterprise-wide viewing of clinical information and provides optimal patient care, it also does Capturing and retrieving discrete patient data, viewing the electronic patient record with the ability to locate discrete results, entering orders, documenting patient care activity, Managing the work day and patient assignments for health care professionals. The Market survey on Powerchart specifies for a particular client on ‘Allergies Documented’ Module on an Average of daily basis around 2560 users are using it. Touchstone 9.1 uses Object Oriented Methodology, where each control on the window is considered as an object. Automated software testing will allow the user to predefine actions, compare the results to the expected behavior and report the success or failure of these manual tests to a test engineer. Once automated tests are created they can easily be repeated and they can be extended to perform tasks impossible with manual testing. Because of this, many companies have found that automated software testing is an essential component of successful development projects. Automated Software Testing saves Time and Money, Improves Accuracy, Increases Test Coverage and Does What Manual Testing Cannot. Example: For an Associate 50 test plans in Manual testing it takes 4 weeks for completing whereas by automating the same it takes only 2 weeks by this there is a huge time line lifts.
Big Data for Personalized Healthcare in Defence
(2018-06-19T11:24:04Z) NAIMA, HUSSAIN; SRUTHI S, NAIR
Effective patient queue management to minimize patient wait delays and patient overcrowding is one of the major challenges faced by hospitals. Unnecessary and annoying waits for long periods result in substantial human resource and time wastage and increase the frustration endured by patients. For each patient in the queue, the total treatment time of all the patients before him is the time that he must wait. It would be convenient and preferable if the patients could receive the most efficient treatment plan and know the predicted waiting time through a mobile application that updates in real time. Therefore, we propose a Patient Treatment Time Prediction (PTTP) algorithm to predict the waiting time for each treatment for a patient. We use realistic patient data from various hospitals to obtain a patient treatment time model for each task. Based on this large-scale, realistic dataset, the treatment time for each patient in the current queue of each task is predicted. Based on the predicted waiting time, a Hospital Queuing-Recommendation (HQR) system is developed. HQR calculates and predicts an efficiency and convenient treatment plan recommended for the patient. Because of the large-scale, realistic dataset and the requirement for real-time response, the PTTP algorithm and HQR system mandate efficiency and low-latency response. We use an Apache Spark-based cloud implementation at the National Supercomputing Center in Changsha to achieve the aforementioned goals. Extensive experimentation and simulation results demonstrate the effectiveness and applicability of our proposed model to recommend an effective treatment plan for patients to minimize their wait times in hospitals.
Body Sensing Device with Authentication and Security
(2018-06-19T10:14:24Z) SUPRIYA, P; UMESH, NAIK; VENKATESH, R; SANTHOSH PRABHU, P
Advances in information and communication technologies have led to the emergence of Internet of Things (IoT). In the modern health care environment, the usage of IoT technologies brings convenience of physicians and patients, since they are applied to various medical areas (such as real-time monitoring, patient information management, and healthcare management). The body sensor network (BSN) technology is one of the core technologies of IoT developments in healthcare system, where a patient can be monitored using a collection of tiny-powered and lightweight wireless sensor nodes. However, the development of this new technology in healthcare applications without considering security makes patient privacy vulnerable. In this paper, at first, we highlight the major security requirements in BSN-based modern healthcare system. Subsequently, we propose a secure IoT-based healthcare system using BSN, called BSN-Care, which can efficiently accomplish those requirements.
Content Based Image Retrieval System
(2018-06-19T09:59:25Z) ROHAN BABU, M; SHAKTHIVELU, A; ALLABAKSH, KARAJGI; RIYAZ, K
Digital image processing is the use of computer algorithms to perform image processing on digital images. As a subcategory or field of digital signal processing, digital image processing has many advantages over analog image processing. It allows a much wider range of algorithms to be applied to the input data and can avoid problems such as the build-up of noise and signal distortion during processing. Since images are defined over two dimensions digital image processing may be model in the form of multidimensional systems. Applications using the images require that the noise be removed making noise removal an important aspect in the image processing. There exist a number of filtering techniques that deal with the noise removal that are included in image enhancement - a preprocessing stage in DIP, some of these may be nonlinear filters. They are usually implemented in software that tends to become slow with increasing image sizes and the bit depths. This project presents "Content Based Image Retrieval". Content-based image retrieval (CBIR), is the application of computer vision techniques to the image retrieval problem, that is, the problem of searching for digital images in large databases. Images are given as input instead of text for efficient way of searching and this reduces time consumption. Applicable to search engines of different fields like marketing, medical etc. The software versions of the algorithms are developed in MATLAB to perform the searching and retrieving the images. MATLAB is a high-level language for numerical computation, visualization and application development and it also provides an interactive environment for iterative exploration, design and problem solving.
Crowd Density Determination for Efficient Public Transportation Facilities
(2018-06-19T08:57:40Z) AYUSH, THAKUR; MARUTI, H; JYOTI, R; JUHISHA, UDHAY U
Existing information systems for urban public transportation are empowering travellers to optimize their trips with respect to travel duration. Experience with such systems shows that this is a viable approach. However, we argue that solely relying on trip duration as the primary indicator for satisfaction can be limiting. Especially, in urban settings providing additional information such as the expected number of passengers can be highly beneficial since it enables travellers to further optimize their comfort. As technical basis for determining the number of passengers, we have built an inexpensive hard-and software system to estimate the current number of passengers in a vehicle. Furthermore, we have deployed the system in several buses in the city of Madrid. In this paper, we describe the overall design rationale, the resulting system architecture as well as the underlying algorithms. Furthermore, we provide an initial report on the system's performance. The initial results indicate that the system can indeed provide a reasonable estimate without requiring any manual intervention.
Cryptography Assisted Key Distribution Using Cloud Computing
(2018-06-19T10:48:34Z) ASHA, YADAV; DEBANJANA, DEY; SABNAM, PANDIT; KRITI, ARYAL
Cloud computing it is the way to stores the data. It provides the data Storage at the high cost and at the faster rate as data is generated. It is costly among the persons who are using it as alone as the Hardware is costly.It help in building up a system that ensures promising environment providing security by implementing the techniques of cryptography and cloud computing. Applying the key distribution approaches during communication and at the same time protecting both security and efficiency is sometimes difficult.Protecting personal privacy and maintaining anonymity is also a major concern in the project.The cloud helps us to reduce the maintaining cost of the storage of data. Cloud storage moves large set of data from the users and the remotely located users.As we know cloud provides the security then it has to face the challenges imposed by the security and resolve them in the better manner. We provide a scheme which gives a proof of data integrity in the cloud which the customer can employ to check the correctness of his data in the cloud. This proof can be agreed upon by both the cloud and the customer and can be incorporated in the Service level agreement (SLA). We apply the methodology of Identity-Based Encapsulation Mechanism to establish communication using anonymous key distribution.“ICC” is an application it provides the security to data in the higher level of data.
Design and Implementation of ATM Security using Finger Print Recognition
(2018-06-19T07:11:53Z) VINDHYA, B V; LAKSHMI, M; KAVYA, M
Identification and verification of a person today is a common thing; which may include door-lock system, safe box and vehicle control or even at accessing bank accounts via ATM, etc which is necessary for securing personal information. The conventional methods like ID card verification or signature does not provide perfection and reliability. The systems employed at these places must be fast enough and robust too. Use of the ATM (Automatic Teller Machine) which provides customers with the convenient banknote trading is facing a new challenge to carry on the valid identity to the customer.Since, in conventional identification methods with ATM, criminal cases are increasing making financial losses to customers. This Paper designs a simple fingerprint recognition system using microcontroller. The system uses R305 fingerprint module to capture fingerprints with its DSP processor and optical sensor. This system can be employed at any application with enhanced security because of the uniqueness of fingerprints. It is convenient due to its low power requirement and portability. The main objective of this system is to develop a system that will increase the ATM security.
Detection and Prevention of Dos Attack On Websites
(2018-06-19T11:37:37Z) SUCHITHRA, A B; RASHMI, D S; KRUTHIKA, P
The web is a complicated graph, with millions of websites interlinked together. In this project, we propose to use this website graph structure to mitigate flooding attacks on a website, using a new web referral architecture for privileged service. This proposed scheme allows a legitimate client to obtain a privilege URL through a click on a referral hyperlink, from a website trusted by the target website. Using that URL, the client can get privileged access to the target website in a manner that is far less vulnerable to a DDoS flooding attack. The proposed solution does not require changes to web client software and is extremely lightweight for referrer websites, which eases its deployment. The massive scale of the web site graph could deter attempts to isolate a website through blocking all referrers.
Effective Method for Removing Intrusive and Incurs overhead by Map Reduce Program
(2018-06-19T11:27:16Z) RAHUL, Y; SABIN, PANDEY; CHETHAN V, MUTTAGI; SUNIL, SINGH
Apache Hive is a widely used data warehousing and analysis tool. Developers write SQL like HIVE queries, which are converted into Map Reduce programs to run on a cluster. Despite its popularity, there is little research on performance comparison and diagnosis. Part of the reason is that instrumentation techniques used to monitor execution cannot be applied to intermediate Map Reduce code generated from Hive query. Because the generated Map Reduce code is hidden from developers, run time logs are the only places a developer can get a glimpse of the actual execution. Having an automatic tool to extract information and to generate report from logs is essential to understand the query execution behavior. We designed a tool to build the execution profile of individual Hive queries by extracting information from HIVE and Hadoop logs. The profile consists of detailed information about Map Reduce jobs, tasks and attempts belonging to a query. It is stored as a JSON document in Mongo DB and can be retrieved to generate reports in charts or tables. We have run several experiments on AWS with TPC-H data sets and queries to demonstrate that our profiling tool is able to assist developers in comparing HIVE queries written in different formats, running on different data sets and configured with different parameters. It is also able to compare tasks/attempts within the same job to diagnose performance issues.
Enhanced Cloud Security using Multilevel Mechanism
(2018-06-19T10:20:39Z) ATISH, OJHA; NIKI KUMAR SAH, KALWAR; SANGYAL, TSERING; SASWATA, CHANDRA
Cloud computing has made an impact on the IT business by proving itself to be the role model for the next generation IT architecture. In divergence to the conventional solutions, where the IT facilities are under appropriate physical and staff office controls, cloud computing provides the flexibility of migrating the system software and databases to the remote data centres, where the management of the data and services may not be fully trustable. This unparalleled attribute, however, poses a good deal of new security challenges which have not been properly analysed. In this project, we specialize in cloud data storage security measures, which has always been an important factor when it comes to quality of service. To ensure the rightness of user's data in the cloud, we devise an efficient and flexible distributed scheme with two prominent features, opposing the conventional methods. By employing three levels of security which involves IP triggering, IP binding, and successful redirection, our schema prevents the misuse of users’ data on the cloud. Unlike most prior works, the new scheme further supports secure and efficient dynamic operations on chunks of data which includes: updating the data, deleting as well as append operations on data. Comprehensive security and performance analysis shows that the proposed scheme is highly efficient and provides an additional layer of security by the method of binding a file to the user’s IP address which prevents data modification attacks from unauthorized addresses and successful redirection to a fake file if in case the credentials are compromised.
Extracting Keywords from Data Set and Assigning Priorities For Popularity Analysis Using Classifier Algorithm
(2018-06-19T10:07:22Z) SAHANA P, REDDY; GAGAN REDDY, N; GAGAN VINAY, HEGDE
Sentiment analysis or opinion mining is one of the major tasks of NLP (Natural Language Processing). Sentiment analysis has gained much attention in recent years. Sentiment is an attitude, thought, or judgment prompted by feeling. Sentiment analysis which is also known as opinion mining, studies people’s sentiments towards certain entities. The main aim is to tackle the problem of sentiment polarity categorization, which is one of the fundamental problems of sentiment analysis. Given a piece of written text, the problem is to categorize the text into one specific sentiment polarity, positive or negative. A general process for sentiment polarity categorization is proposed with detailed process descriptions. Data used in this study are college reviews collected from unigo.com. Experiments for both sentence-level categorization and review-level categorization are performed with promising outcomes. However the data have several flaws that potentially hinder the process of sentiment analysis. The first flaw is that, since people can freely post their own content, the quality of their opinions cannot be guaranteed. The second flaw is that, the ground truth of such data is not always available. A ground truth is more like a tag of certain opinion, indicating whether the opinion is positive, negative, or neutral. There are three levels of sentiment polarity categorization, namely the document level, the sentence level, and the entity and aspect level. All the sentences were firstly tokenized into separated English words. The syntactic roles are also known as the parts of speech. In natural language processing, parts-of-speech (POS) taggers have been developed to classify words based on their parts of speech. The second process involves the sentiment score computation for the sentiment tokens. The sentiment score depicts the level of positivity or negativity of a particular sentiment word. Sentiment tokens and sentiment scores are information extracted from the original dataset. Then ratings for the colleges can be made by using those polarities. The efficiency or accuracy that can be achieved using sentiment analysis is about 60-70%
An Eye for the Visually Impaired to Know the Bus Number and The Type of Bus Arriving at the Bus Stop
(2018-06-19T07:06:45Z) NAVEEN, RAJ; RUDRA, ROYCHOUDHURY; KAPIL, KUMAR; MANISH
At the bus stops we usually see visually impaired people asking help of people present over there to know the bus number of the buses arriving . They have to depend on others for knowing the information and details related to the buses arriving. This information makes them to decide whether or not to board the bus as per their destination. The required information are generally the bus number, its destination and the type of bus(like Volvo,KIA,etc).This project is designed to make the blind independent of others and not asking for help. This will help the visually impaired at the bus stops to automatically know the details of the next bus arriving and as per that information they can instantly come up with a decision of boarding the particular bus. We will use a RFID cards and card reader to track the bus which will help in knowing the bus number and destination. Image processing is used to identify colour of the bus, which helps in knowing the type of bus. E.g. Volvo buses have Blue colour, KIA buses are green in colour, in normal BMTC buses also there are various types. This above information will also help in knowing the number of doors the bus has, some buses have single door and some have two doors. Accordingly, visually impaired can stand to get into the bus. The information after image processing will be converted into speech and sent to the required bus stops via speakers as output.
Eye State Detection of Driver
(2018-06-19T11:34:48Z) BLESSINA D, JUTIKE; CHAITANYA, K; JYOTHI, M
Eye state detection of a driver system is an effective tool to reduce the number of road accidents. This project proposes a non-intrusive approach for detecting drowsiness in drivers, using Computer Vision. The algorithm is coded on OpenCV platform in Windows environment. The parameters considered to detect drowsiness are eye detection, blinking, eye closure and gaze. The input is captured from the camera and live fed the algorithm is Haar Classifier trained to detect the face and the eye from the incoming frame. Once the eye is detected,to track the eye and automatically set a dynamic threshold value. Depending on the values obtained from each of the incoming frames and deviations from the threshold values, eyelid closure/blink/gaze is detected. Warning system is designed to alert the driver. This system renders an efficient solution to road accidents and the cost of developing it into a real time system is also feasible when compared to the cost involved in the manufacture of car.
Fast and Efficient Data Search in Hadoop
(2018-06-19T10:32:01Z) T.POOJITHA, REDDY; MANDARA, B M; NANDITHA, N
Hadoop is the most popular implementation framework of the MapReduce programming model, and it has a number of performance-critical configuration parameters. However, manually setting these parameters to their optimal values not only needs in-depth knowledge on Hadoop as well as the job itself, but also requires a large amount of time and efforts. Automatic approaches have therefore been proposed. Their usage, however, is still quite limited due to the intolerably long searching time. In this proposed system, we introduce MapreducE Self-Adjusting (MESA), a framework that accelerates the searching process for the optimal configuration of a given Hadoop application. We have devised a novel mechanism by integrating the model trees algorithm with the genetic algorithm. As such, MESA significantly reduces the searching time by removing unnecessary profiling, modelling, and searching steps, which are mandatory for existing approaches. Our experiments using five benchmarks, each with two input data sets (DS1 and 2xDS1) show that MESA improves the searching efficiency (SE) by factors of 1.37x and 2.18x on average respectively over the state-of-the-art approach.
Fault Prediction of an Actuator or Sensor of a Satellite
(2018-06-19T09:12:37Z) GIREESH GANAPATI, GAONKAR; GOKUL VIHARI, KODURU; H.S., SUMANTH; K.S., SHASHANK
This paper will fortify the existing hardware by predicting the fault in the reaction wheel. Reaction wheels present in the AOCS may fail over the time, mainly due to friction. We collect the data from satellite and use bigdata analytics and statistical measures to extract the new insights and regression analysis is performed which is a part of machine learning and predict behaviour of the parameters. We implement this in MATLAB and RStudio. Computation process is carried out in cloud and amazon AWS (ec2 instance) is used for this purpose. If the fault is predicted warning is sent about the time of failure.
Gesture Recognition System
(2018-06-19T09:42:05Z) MUHAMMED, NISAMUDHEEN; MUTHUMARIYAPPAN, U; SUMANTH, S; YOGEESH G, KONNUR
The project is about a real time Hand Gesture Recognition and feature extraction using a web camera. • In this approach, the image is captured through webcam. • Image from the webcam is preprocessed and threshold is used to remove noise from image and smoothen the image. • When the preprocessing is complete the image is passed on to feature extraction phase. • The extracted features are normalized and matched with the training dataset features using KNN (K-nearest neighbor) algorithm (Euclidean distance).
Handwritten Document Conversion Using Regional Convolutional Neural Network
(2018-06-19T09:38:51Z) MOHAN KRISHNA, B; SWATHI, MADHAVAN; PRAKASH, H; RAHUL, N
This project converts handwritten document into its corresponding digital format using Regional Convolutional Neural Network (R-CNN). R-CNN is the pre-eminent machine learning visual object detection algorithm. Since, digital documents can be easily manipulated, stored and retrieved, this project serves various applications such as processing cheques in banks, converting handwritten books into digital copies that can used for publication, retrieving information from application forms, etc. Firstly, the handwritten document image is pre-processed to produce better results using adaptive thresholding, which removes noise and artifacts in the image. Secondly, character segmentation is performed in order to obtain individual handwritten characters from the image. The position of each of these characters is retrieved in order to position the characters in the final text document. Thirdly, these characters are passed one-by-one into the Convolutional Neural Network (CNN) model, which performs classification and produces the probability of each character class with one character class having the highest probability using softmax function. The character with the highest-class probability is chosen as the output character and is written into the text document based on its corresponding position.
Implementation and Analysis of Machine Learning on Cyber Intrusion Detection
(2018-06-19T09:51:43Z) PRABHAT KUMAR, PRASAD
The focus of the project is, implementation of Machine Learning models to detect intrusions in a computer network from unauthorized users, including perhaps insiders. The task of the intrusion detector is to build a predictive model capable of distinguishing between intrusions or normal connections. A connection is a sequence of packets flowing to and fro, a source IP address to a destination IP address under some well-defined communication protocol. Each connection is labelled as either normal, or as an attack, with exactly one specific attack type. Attacks fall into four main categories: 1. DOS: denial-of-service, e.g. syn flood; 2. R2L: unauthorized access from a remote machine, e.g. guessing password; 3. U2R: illegitimate access to local super user (root) perquisite, e.g., various ``buffer overflow'' attacks; 4. Probing: surveillance, traffic analysis and other probing, e.g., port scanning. It's important to note that the probability distribution of test data is not from the same as the training data, and it comprises of specific attack types not in the training data, making the task more realistic. The attack types in training dataset and test dataset are total of 24, and additional 14 respectively. Experts in this field believe that majority of the novel attacks are variants of known attacks and the "signature" of known attacks can be sufficient to catch novel variants. With the following ideas as base for the project, we will experiment with different machine learning approaches. This project sheds light on complexities, peculiarities and potential of using ML Algorithms for Cyber Security Intrusion Detection and the results of these models are deeply analysed with various evaluation strategies, and also via visualisations.

Browse

Browsing 2017-18 by Title

Results Per Page

Sort Options