2018-19

Permanent URI for this collection

http://172.16.1.132:4000/handle/123456789/10717

Browse

Now showing 1 - 20 of 48

Restorehealth- Prescription based Healthcare Services
(2019-06-25T12:45:29Z) Jyotishmita Patangia
The project presents a novel method of collaborating ease in purchase of medicines through online shopping in the sense of security moneywise as well as for customer satisfaction while doing shopping offline. This is implemented using a windows application. In offline mode the customer needs to physically pick up his purchase and in online mode is doorstep delivery providing methods to edit the cart. Along with these it keeps a record of the customers health by providing routines of medicines and alert messages too. Healthcare services provide a wide range of prescription medicines and other health products conveniently available all across India. Even second and third tier cities and rural villages can now have access to the latest medicines. Since we also offer generic alternatives to most medicines, online buyers can expect significant savings. We not only provide you with a wide range of medicines listed under various categories, we also offer a wide choice of OTC products including wellness products, vitamins, diet/fitness supplements, herbal products, pain relievers, diabetic care kits, baby/mother care products, beauty care products and surgical supplies.
CREDIT CARD FRAUD DETECTION
(2019-06-25T12:42:59Z) ASHWINI H V
Nowadays the usage of credit cards has dramatically increased. As credit card becomes the most popular mode of payment for both online as well as regular purchase, cases of fraud associated with it are also rising. Here we model the sequence of operations in credit card transaction processing using a Hidden Markov Model (HMM) and show how it can be used for the detection of frauds. An HMM is initially trained with the nonnal behavior of a cardholder. If an incoming credit card transaction is not accepted by the trained HMM with sufficiently high probability, it is considered to be fraudulent. At the same time, we try to ensure that genuine transactions are not rejected. We present experimental results to show the effectiveness of our approach and compare it with other techniques available in the literature.
TDOA based Speaker Localization using MATLAB
(2019-06-25T12:40:51Z) Vicky Chetri
Source localization has always been a challenging problem wherein the main focus is to give an estimate about the coordinates of different sound sources that generates a signal which is then received by a pair of microphones. In this process we have considered a Time Delay of Arrival based source localization which uses a wireless sensor network. In this we show that the association ambiguity of TDOAs can be effectively resolved using the concept of an inverse delay interval region (IDIR). —Multiple source localization (MSL) using time differences of arrival (TDOAs) is challenging because of the ambiguity involved in associating the TDOAs computed across microphone pairs to the sources. We show that the association ambiguity of TDOAs can be effectively resolved using the concept of an inverse delay interval region (IDIR), which we introduce in this paper. By examining the association between a spatial domain and the TDOAs, we define IDIR as an inter-hyper boloidal spatial region corresponding to an interval of delays for a given pair of microphones. The proposed system for localizing multiple sources of sound involves two stages that is in the first stage, the given waves is partitioned into non-overlapping elemental regions and which contains a source are then detected by measure based on the generalized cross-correlation with phase transform (GCC-PHAT), and the IDIRs. In the second stage, we see that the sources are finely localized within each of the detected regions by identifying
Bridge Monitoring and Alert Generation System
(2019-06-25T12:37:04Z) Varshini A
Many of the bridges in cities built on the river are subject to deterioration as their lifetime is expired but they are still in use. They are dangerous to bridge users. Due to heavy load of vehicles, high water level or pressure, heavy rains these bridges may get collapse which in turn leads to disaster. So, these bridges require continuous monitoring. So we are proposing a system which consists of a weight sensor, water level point contact sensor, Wi-Fi module, and Arduino microcontroller. This system detects the load of vehicles, water level, and pressure. If the water level, water pressure and vehicle load on the bridge cross its threshold value then it generates the alert through buzzer and auto barrier. If it is necessary, then the admin assign the task to the employees for maintenance.
Multi-Banking Transaction ATM System using Biometric and GSM Authentication
(2019-06-25T12:33:27Z) THANUSHA REDDY J
Frauds attacking the automated teller machine has increased over the decade which has motivated us to use the biometrics for personal identification to procure high level of security and accuracy This paper describes a system that replaces the ATM cards and PINs by the physiological biometric fingerprint and iris authentication. Moreover, the feature of one-time password (OTP) imparts privacy to the users and emancipates him/her from recalling PINs. Additionally, the system provides protection to the ATM terminal from fire and thief attacks by making provisions of pump motor and a DC motor for rolling the shutter. In this system during enrolment the genuine user’s fingerprint samples of are retained in the database. The procedure of exchange starts by catching and coordinating fingerprints. The framework will consequently recognize genuine authentic characteristic and phony examples. A GSM module connected to the Arduino will message a code generated by the system to the registered mobile number. After the substantial OTP is entered the client can either pull back or store money or check his/her parity. In any sort of phony access endeavor’s, the record is hindered. In this paper the test results are gotten on the informational index of unique finger impression progressively utilizing unique mark module with a particular coordinating calculation and a GUI based of Round Hough Change separately.
CHATTERBOT USING DEEP LEARNING
(2019-06-25T12:30:01Z) SUSHREE SANGEETA DALABEHERA
Chatbots are replacing some of the jobs that are traditionally performed by human workers, such as online customer service agents and educators. From the initial stage of rule-based chatbots to the era of rapid development in artificial intelligence (AI), the performance of chatbots keeps improving. Chatbots can nowadays “chat” like a human being and they can learn from experience. The purpose of this research is to examine the past research on chatbots (also known as conversational agents) using the quantitative bibliometric analysis. The contribution of this research is to help researchers to identify research gaps for the future research agenda in chatbots. The results of the analysis found a potential research opportunity in chatbots due to the emergence of the deep learning technology. This new technology may change the direction of future research in chatbots. Several recommendations for future research are provided based on the results obtained from our analysis.
AGRICULTURAL INTELLIGENCE DECISION SYSTEM
(2019-06-25T12:26:57Z) SURAJ SHET H
In the recent years, the huge volume of real time data in the agricultural sector and its need for an efficient and effective processing, stimulate the use of novel technologies and platform to acquire, store, process, analyze and visualize large data sets for future predictions and decision making. Big Data is an evolving term given to a wide area of data-intensive technologies in which the datasets are extremely large that dealing with them become more challenging than how it was. Before, Due to the critical challenges facing the agriculture sector farmers feel more forced to adopt intensive farming practices and sustainable agricultural ones, in order to increase both economic and environmental costs. Bringing data mining technologies into agriculture presents a significant challenge; at the same time, this technology contributes effectively in many countries’ economic and social development. In this work, we will study environmental data provided by precision agriculture information technologies, which represents a crucial source of data in need of being wisely managed and analyzed with appropriate methods and tools in order to extract the meaningful information. i
A multi Modal Deep Learning Method for Android Malware Detection Using Features
(2019-06-25T11:29:33Z) Sujeet Kumar
With the widespread use of smartphones, the number of malwares has been increasing exponentially. Among smart devices, Android devices are the most targeted devices by malware because of their high popularity. This paper proposes a novel framework for Android malware detection. Our framework uses various kinds of features to reflect the properties of Android applications from various aspects, and the features are refined using our existence-based or similarity-based feature extraction method for effective feature representation on malware detection. Besides, a multimodal deep learning method is proposed to be used as a malware detection model. This paper is the first study of the multimodal deep learning to be used in the Android malware detection. With our detection model, it was possible to maximize the benefits of encompassing multiple feature types. To evaluate the performance, we carried out various experiments with a total of 41,260 samples. We compared the accuracy of our model with that of other deep neural network models. Furthermore, we evaluated our framework in various aspects including the efficiency in model updates, the usefulness of diverse features, and our feature representation method. In addition, we compared the performance of our framework with those of other existing methods including deep learning-based methods.
Election Sentimental Analysis in iOS
(2019-06-25T11:27:04Z) Sudip Kandel
The process of computationally identifying and categorizing opinions expressed in a piece of text, especially in order to determine whether the user's attitude towards a particular topic, product, etc. is positive, negative, or neutral is called as sentiment analysis. During election days , it’s really necessary to determine the mindset of the people towards particular tweets or sentiment. In this project we used twitter API to fetch the data using the unique key provided by the twitter. An user interface is constructed in iOS platform which can be used to search tweet related to the election . These tweets are later classified using a machine learning model(text classifier) which calculates the points with respect to the number of tweets according to its sentiment (Positive, Negative or Neutral). These points are used to determine the actual sentiment of the people towards that particular search. Later , this classified points are displayed in user’s screen. Users also can make multiple searches as per their requirement
A Three-Layer Privacy Preserving Cloud Storage Scheme Based on Computational Intelligence in Fog Computing
(2019-06-25T11:22:16Z) Sudarshan Thirumalai
Recent years witness the development of cloud computing technology. With the explosive growth of unstructured data, cloud storage technology gets more attention and better development. However, in current storage schema, user’s data is totally stored in cloud servers. In other words, users lose their right of control on data and face privacy leakage risk. Traditional privacy protection schemes are usually based on encryption technology, but these kinds of methods cannot effectively resist attack from the inside of cloud server. In order to solve this problem, we propose a three-layer storage framework based on fog computing. The proposed framework can both take full advantage of cloud storage and protect the privacy of data. Besides, Hash-Solomon code algorithm is designed to divide data into different parts. Then, we can put a small part of data in local machine and fog server in order to protect the privacy. Moreover, based on computational intelligence, this algorithm can compute the distribution proportion stored in cloud, fog, and local machine, respectively. Through the theoretical safety analysis and experimental evaluation, the feasibility of our scheme has been validated, which is really a powerful supplement to existing cloud storage scheme. Key words: Computing Technology, Three-Layer Storage, Cloud Computing, Fog Computing, Encryption Technology, Hash-Solomon Code.
FOOD QUALITY INSPECTOR
(2019-06-25T11:19:48Z) SOWMYA SHREE .N
Food security and food quality are some of the major issues of our civilizations since ages. For conducting sorting and regular quality checks companies employ much manual labor. And wherever we have manual work, efficiency and productivity are some of the major concerns. As a result the economic costs involved in managing this process dent the bottom line of a company. Thus there is a real need to optimize this process through automation. The advancements in machine learning and artificial intelligence now make it possible to replicate the manual visual inspection with the help of machines. Machine learning technology enables us to visually scan the vegetables and classify them into good or bad. Thus the proposed project visually inspects tomatoes using Machine Learning via the Dlib machine learning toolkit to scan tomatoes for performing automated visual inspection of their quality. Keywords: Machine learning, Dlib, Imglab, Opencv, Image processing,Computer vision.
AZURE AD AUTOMATION BOT
(2019-06-25T11:17:04Z) Shyam R
Microsoft Azure is a cloud computing service created by Microsoft for building, testing, deploying, and managing applications and services through Microsoft-managed data centers. Azure Active Directory (Azure AD) is Microsoft's cloud-based identity and access management service that is controlled, authorized or managed by an administrator of an organization. In this project an administrative manual task is being automated. That is, creating a new azure ad account for an user. The whole automation task is done here by using Microsoft Flow. Microsoft Flow is cloud-based software that allows employees to create and automate workflows and tasks across multiple applications and services. Automated workflows are called flows. Luis is a machine learning-based service to build natural language into apps, bots, and IoT devices. To create enterprise-ready, custom models that continuously improve. In this project, a Natural language understanding bot that will understand user request by connecting with LUIS and in turn perform the task by connecting with Microsoft Flow and successfully return back the corresponding result as a message in the bot is created. The bot is developed using JavaScript & NodeJS as the language. Next, a LUIS application that will accordingly understand user utterances in the bot was created and connected to the bot. Later, a workflow in Microsoft Flow was created to automate the tasks done by an IT admin in creating a new azure ad account. Again, the workflows are called to the bot using Microsoft Graph API calls. Finally, when the user sends a request in the bot saying “create an ad for ” or any other requests similar or related to the previous request results in the successful creation of azure ad user account for the respective name mentioned. A success message is sent back immediately in the bot after the successful completion of the task.
Intelligent Canister
(2019-06-25T11:14:24Z) Shubha B N
Many aspects of day to day tasks have been greatly influenced by technological progression. This is especially true in the case of automation as well as data storage and access. With the increasing availability of microcontrollers at reasonable costs, it becomes intuitive to apply such a device to facilitating day to day tasks. The Arduino Uno in specific is one such microcontroller that helps you build vast application products that are helpful for your daily needs. An load sensor, LCD sensor, a large array of general-purpose inputs and outputs, and even a wireless internet module that is a GSM module. Applying these concepts to produce a product such as an automated pantry system with active inventory would provide another step towards completing daily tasks with much greater efficiency. The Intelligent canister is a modular system and can be adjusted to the different consumers needs. At its base is a small (and low price) computer (could be a low end pc or similar). This base can make use of a different combination of modules to keep an updated list of the commodities in your pantry. A weighing station module will enable communication through small weight sensors that can be placed under strategic commodities that will be tracked by weight. Many aspects of day to day tasks have been greatly influenced by technological progression. This is especially true in the case of automation as well as data storage and access. With the increasing availability of microcontrollers at reasonable costs, it becomes intuitive to apply such a device to facilitating day to day tasks.The Arduino Uno in specific is one such microcontroller that allows for simpler display interfacing ultrasonic sensor , LCD sensor, a large array of general-purpose inputs and outputs. Applying these concepts to produce a product such as an automated pantry system with active inventory would provide another step towards completing daily tasks with much greater efficiency. i
F.E.S.R (Fire Extinguishing Safety Robot)
(2019-06-25T10:31:44Z) SHRUTHI SK
It’s been strenuous for fire-fighters to access the site of fire due to high temperature or presence of explosive material which causes damage to life and property. In such environments fire-fighting robots can be of immense help, they can also be used to protect fire-fighters from extreme petrochemical, toxicity or explosive fire accidents .The main aim of this robot is to extinguish the fire using wireless technology. If the path is complicated then it is difficult to stop the fire. So by using F.E.S.R we can overcome such hurdles. The principles used in this design are such that it enables the robot to be extended to a more robust system to be used to combat actual fires in residential or commercial settings. The main requirement of this project is to create a robot that is fully autonomous. This means that once the robot is started by the user, it navigates, searches for, and extinguishes the fire on its own, without any input from the user .In order to reach this goal, we made many critical decisions on motor, sensors, fire extinguishing mechanical parts and general design of the robot. Lastly the robot should not only accomplish these requirements, but also be able to do so quickly and accurately. The brain of the robot is the Arduino ad the fire is detected using flame sensors, it’s an IoT project. Keywords: Wireless technology, IoT(Internet of things), flame sensors
Fresh Recipes Web Application
(2019-06-25T10:29:28Z) SAI PRASHANTHI S
A recipe management web application is developed which can be used for saving and reading the recipes. The present say people want all things automated and digitalised. Olden days people used to store recipes in books and save it. These cookbooks are not very safe and may be damaged and lost. Also looking for recipes from these cookbooks is tedious and remembering which recipe we saw where and which book is difficult. The application consists of various modules including a module to upload the recipes to the database along with image, a home page that consists of all the recipes that are added to the database . It also consists of the login and sign up modules which authenticate the users and customize the view for the authorized users. The application will also consist of a search module where the users can easily search for what they are exactly looking for. This makes it extremely easy for the people to store and use the recipes. Recipe storage becomes digital and time saving. The above modules are implemented using JavaScript frameworks such as node js, express js, passport js, Ajax etc. The front end fully responsive layout is developed using html , bootstrap and css3 Keywords— login authentication, passport js, express js framework ,search module. i
Rating Prediction On Electronic Gadgets
(2019-06-25T10:27:05Z) Sahana C A
Mining feeling targets and assessment words from online audits are significant errands for fine-grained conclusion mining, the key segment of which includes distinguishing supposition relations among words. In this framework a novel methodology dependent on the somewhat administered arrangement model, which views distinguishing feeling relations as an arrangement procedure? At that point, a chart based co-positioning calculation is misused to assess the certainty of every applicant. At last, competitors with higher certainty are separated as sentiment targets or assessment words. Contrasted with past techniques dependent on the closest neighbor administers, our model catches supposition relations all the more correctly, particularly for long-length relations. Contrasted with punctuation based strategies, our pledge arrangement model adequately mitigates the negative impacts of parsing mistakes when managing casual online writings. Specifically, contrasted with the customary unsupervised arrangement model, the proposed model acquires better accuracy in light of the utilization of fractional supervision. Likewise, while assessing applicant certainty, we punish higher-degree vertices in our chart based co-positioning calculation to diminish the likelihood of blunder age.
BITCOIN PRICE PREDICTION
(2019-06-25T10:24:48Z) Raksha Ratnakar Mahale
Bitcoin is a crypto currency which is used worldwide for digital payment or simply for investment purposes. The record of all the transactions, the timestamp data is stored in a place called Blockchain. Each record in a blockchain is called a block. The aim of the project is to understand and identify daily trends in the Bitcoin market. The data set consists of various features relating to the Bitcoin price and payment network. Using the available information, I will predict the sign of the daily price change with highest possible accuracy. The proposed model uses algorithms like Bayesian regression and Generalized linear model (GLM) / Random Forest, which will increase the efficiency with which the price of bitcoin can be predicted more accurately compared to other algorithms. As bitcoin is growing drastically in recent times, it is very important to predict the price of it more accurately as it is going to be future trend of investment. If prices are predicted more accurately then it will be helpful for people to invest on it.
AUTOMATIC BILL GENERATING SYSTEM FOR VEHICLE PARKING
(2019-06-25T10:22:35Z) RAKSHA S R
In this paper we have shown the concept of Microcontroller Based on Automatic Bill Generating System for vehicle parking. As we see in the modern world everything is going automatic we have built a system which will automatically generate bill for vehicle parking system. It includes the concept of Microcontroller for Automatic bill generating system for vehicle Parking. It automatically sense whether the cars are present in the parking lot with the help of microcontroller. It tells about the number of cars in the parking lot. We have deployed a microcontroller which is used to sense the movement of cars and depending upon whether there is a capacity of cars to enter, it either opens the gate or not. There is also RFID module to provide security as users who have authority can swap the RFID cards and get entry otherwise not. The project is designed for car parking. The goal of this paper is to automatically park the car for allowing the cars into the parking area. LCD display is provided to display the information about the number of cars that can be parked and the place free for parking. Keywords: Microcontroller, Parking System, IR Sensors, LCD, LED, RFID Tag, Servo motor.
SKIN DISEASE DETECTION USING CONVOLUTIONAL NEURAL NETWORK
(2019-06-25T10:20:21Z) PUJA
Skin diseases are very common in people’s daily life. Each year, millions of people are affected by all kinds of skin disorders. Diagnosis of skin diseases sometimes requires a high-level of expertise due to the variety of their visual aspects. As human judgment are often subjective and hardly reproducible, to achieve a more objective and reliable diagnosis, a computer aided diagnostic system should be considered. In this project , we investigate the feasibility of constructing a universal skin disease diagnosis system using deep convolutional neural network (CNN). The key part of architecture is a Convolution Neural Network that is trained on a skin disease image database. The dataset is obtained from skin disease database available openly HAM10000 dataset. Seven classes of diseases are predicted. It uses softmax layer of CNN for disease prediction. Our project can achieve as high as 90% accuracy. The accuracy can be further improved if more training images are used.
IMPROVING VIDEO CLASSIFICATION ACCURACY USING CLOUD
(2019-06-25T10:14:18Z) Prabin Mandal
The focus of this project is on the frame level features. One of the promising algorithms that can be used for this purpose is Deep Bag of Frame pooling (DBoF). Deep bag of frame model is a convolutional neural network (CNN). The main idea is to design two layers in the convolutional part. The approach enjoys the computational benefits of CNN, while at the same time the weights on the up-projection layer can still provide a strong representation of input features on frame level. The classification is performed at the final layer of the CNN. We will use the Youtube-8M dataset for experimentation. The Youtube-8M dataset is the largest publicly available multi-label video classification dataset, with approximately 8 Million videos annotated with 3862 classes of labels. The videos within the dataset averages 3.01 labels per video, where the number of labels per video ranges from 1 to 23. As this dataset covers over 500,000 hours of video, 2.6 billion audio and visual features have been extracted and pre-processed in advance by the Google Research Team as it would be infeasible for research teams to train hundreds of Terabytes worth of video for their mode.

Browse

Recent Submissions