About admin

This author has not yet filled in any details.
So far admin has created 790 blog entries.


Bhoomi Shah 1,Shubham Singhal2  

1Department of Computer Engineering MIT College Of Engineering Pune, India  

2Department of Computer Engineering MIT College Of Engineering Pune, India 

ABSTRACT:  Pattern recognition of hand gestures associated with different alphabets and key press events contributes enormously towards human computer interaction. Nonverbal communication using hand gesture provides user friendly way of interaction with the computer. A recognition system is developed to translate hand gestures into key press events using only shape parameters. Real time system for hand gesture recognition is largely based on detection of meaningful shape based features. This paper is a survey on gesture recognition approaches in which includes static hand posture methods that are explained with different tools and algorithms applied on gesture recognition system namely hidden Markov model and fuzzy clustering.

Keywords: PUI, kinesis, k-means clustering, HMM, fuzzy c logic, threshold



By | 2017-07-06T10:08:13+00:00 July 6th, 2017|Conference Publications|0 Comments


Prof. Dr.Prachi Joshi1, Vinit Bhosale1, Ketan Kodmelwar1, Ninad Naik1, Komal Ramole1

1Department of Computer Engineering, MIT College of Engineering, Pune 

ABSTRACT:  In recent years, interactions between machine and humans has become an issue of concern. Speech is most natural form of thought in which human communicate. Speech recognition involves transformation of speech signal into sequence of words with help of algorithm. Analysis of speech can lead to capture the certain important aspects which can help in determining individual characteristics. This further can also help in capturing grasping power and detecting speech disabilities of children with special needs. This paper gives a survey of techniques used for speech processing.

Keywords: Communication, Speech, KNN, SVM, Machine Learning, Speech features



By | 2017-07-06T10:05:43+00:00 July 6th, 2017|Conference Publications|0 Comments


Prof. Sumitra Pundlik 1, Pritam Rikame2, Neeraj Upadhyay3, Mayuresh Kakade4,  

Yash Tatkondawar  5  1Department of Computer Engineering   2Department of Computer Engineering MIT College of Engineering, Pune, India 

ABSTRACT:  Recently there has been drastic change in how people take care of their health. So the modern health care system must be capable of identifying any diseases symptoms that might appear in the individual. For detecting these symptoms modern system makes use of different signals that were collected from the individual under test. Further processing of these signals are done and finally machine learning algorithm are applied for predicting the final result.

Keywords: Ayurveda, Wrist pulse analysis, Tridosha, ANN, SVM, Random Forest Algorithm



By | 2017-07-06T10:01:49+00:00 July 6th, 2017|Conference Publications|0 Comments

Prof. Sumitra Pundlik 1,Vedashree Patil2, Sejal Upadhye3, Chaitanya Ujeniya4, Rajul Badal  5

 1Department of Computer Engineering   2Department of Computer Engineering MIT College of Engineering, Pune, India

 ABSTRACT: Deep learning is the newest field of machine learning that is gaining widespread popularity in almost all fields and has piqued the interest of all major technology giants including the likes of Google, Facebook, Amazon etc. With the availability of large volumes of data-sets along with the widespread use of GPUs that make parallel processing faster, cheaper and more powerful, machines can learn representations of image, text and audio and produce abstractions with the deep learning approach with lower error rates.

Keywords: Deep learning, GPU, Computer Vision, Object detection and recognition 




By | 2017-07-06T09:43:38+00:00 July 6th, 2017|Conference Publications|0 Comments



1Prof.Anita Thengade,2Anurag Bohra

 1,2Department of Computer Engineering, MIT College of Engineering, Pune, India 


Computers have become more pervasive and linked to Internet, they are open to a cornucopia of security threats ranging from unsolicited e-mail messages to drive-by downloads to botnets. These threats have become more vulnerable in recent years and the conventional methods used till now have become obsolete. Machine Learning is considered to be an ideal solution for this issue. This report considers all the aspects of integrating machine learning and computer security and based on those aspects and facts conclude whether it causes an advancement in securing computers or deteriorate the same.

Keywords: Computer Security, Machine Learning, Supervised Systems



By | 2017-07-06T09:39:32+00:00 July 6th, 2017|Conference Publications|0 Comments


Sarvesh Tendulkar, Swarada Adavadkar

Department of Computer Engineering   MIT College of Engineering Pune, India 


Data Mining (DM) is extracting information from large set of data and obtaining meaningful results from it. Data Mining has evolved into diverse fields which requires development of data collection, creation, data management (including data storage retrieval and database transaction processing) and advanced data analysis (data warehousing and mining).This paper explains about two of Data Mining fields : Educational Data Mining (EDM) and Web Data Mining (WDM). Educational Data Mining (EDM) is an emerging field of data mining which applies various data mining algorithms for betterment of students and improving quality of student courseware. Web Data Mining (WDM) is a technique used to crawl through various web resources to collect information which an individual or a company use to promote business. In this survey, different published research papers on methodologies used in EDM and WDM are compared. Also, it gives an insight into techniques used in EDM and WDM.

Keywords: Keyword Indexing, Search engines, document frequency, term weights, data mining, neural network, predicting school failure, job trends analysis




By | 2017-07-06T09:36:56+00:00 July 6th, 2017|Conference Publications|0 Comments


Jayshree Ghorpade1, Pratik Kataria2, Prachi Dwivedi 3, Kiran Rode 4, Akshay Jain 5

Department of Computer Engineering, Maeer’s MIT-COE, Pune, India

ABSTRACT: JavaScript has advanced significantly as an integrated technology stack for building user interfacing applications on the front-end as well as the back-end. There is a school of thought that the future of JavaScript is so tightly coupled to the web that it must exist if the client side web does. JavaScript is widely used. It can be used for Full-Stack Development and many other languages or supersets can be compiled into JavaScript. This paper is written to explore the idea of using JavaScript for Small Data as well as Big Data platforms. The front-end as well as the back-end is explained to infer whether the concept of ‘JavaScript Everywhere’ is actually possible in the future. Thus taking into consideration the past and present of JavaScript, the future of JavaScript is anticipated. If there’s a browser, then there has to be JavaScript associated with it. It may happen that JavaScript is also used for kitchen appliances or home amenities as a whole for IoT.

Keywords : Data analysis, Data visualisation, JavaScript, D3.js, front-end, back-end, JSON



By | 2017-07-06T09:33:42+00:00 July 6th, 2017|Conference Publications|0 Comments


Prof. Anita Gunjal1, Akshata Talele1, Chinmayee Mundhe1,  Nikita Adkar1 , Manodnya Bhoite1

1MAEER’s MIT College of Engineering, Pune 

Abstract: When Internet first came to the world it was considered a revolution and now we are at a precipice of the applications of Internet and time has come when we take a leap to a new revolution and that is the connection of people and processes i.e. Internet of Things (IoT). It is a world where physical objects are integrated in the network to share information. Over the last few years, the development in IoT enabled the fascinating applications. The aging population is constantly rising worldwide, both in urban and regional areas. Hence, there is a need of IoT based remote healthcare monitoring. These systems can take care of the elderly people without compromising their preferences and convenience. In this paper, we identified the network requirements of remote healthcare monitoring systems in the terms of bandwidth requirements, data generation and real-time event update. Further we studied about the protocols such as XMPP, HTTP, CoAP and MQTT.

Keywords : Internet of Things, Remote healthcare monitoring system, Protoc


By | 2017-07-06T09:29:19+00:00 July 6th, 2017|Conference Publications|0 Comments


Aditya Deshpande1, Alisha Shahane 2, Darshana Gadre3, Mrunmayi Deshpande4, Prof. Dr. Prachi M. Joshi

1Department of Computer Engineering, MAEER’s MIT College of Engineering, Pune, India 

Abstract: Today is the era of intelligence in machines. With the advances in Artificial Intelligence, machines have started to impersonate different human traits today. Artificial intelligence conversational entities, also called chatbots, are an excellent example of such machines. Chatbots are computer programs capable to carry out near – natural conversation with people. In this work, we describe the evolution of chatbots from a rudimentary model to an advanced intelligent system. Chatbots are currently gaining a lot of popularity especially in business sector as they have the potential to automate customer service and reduce human efforts. For a chatbot to perfectly emulate a human dialogue, it must analyze the input given by a user correctly and formulate a relevant and appropriate response.

Keywords: Chatbots, OCR, Information Repositiory, Knowledge Base, AI


By | 2017-07-06T09:25:56+00:00 July 6th, 2017|Conference Publications|0 Comments


Sumeet Ghegade 1, Mandar Koujalgi 2, Garvit Raj Singhal3 

 1Department of Computer Engineering   2Department of Computer Engineering MITCOE, India 


Stock market trading is becoming a major source of income for people in India. In order to be a successful investor the most important thing is to predict the trend of the stock you are investing in. It is a well-known fact though that stock market prices are highly volatile. Analyzing the entire data and recent updates about a company is beyond human brains comprehension. Hence, we aim to propose in this paper some models that may be able to predict accurate trends in stock market prices. The aim is to forecast closing price of a particular stock for next few days with minimal error. Combination of both technical and fundamental analysis is going to be used along with some data analysis techniques.

Keywords: Stock market, trends, fundamental analysis, technical analysis, data analysis techniques.



By | 2017-07-06T09:22:02+00:00 July 6th, 2017|Conference Publications|0 Comments


Mohammad Ashraful Islam1 Md. Anin Naeem2, Md. Nazmul Hasan2

1 Lecturer, Department of Computer Science and Engineering, Mawlana Bhashani Science & Technology University, Bangladesh

Student, Department of Computer Science and Engineering, Mawlana Bhashani Science & Technology University, Bangladesh

Face detection is the first step of automated face recognition process in all pattern matching or artificial intelligence or robotic vision system. There are many face detection algorithms and they have their own strengths and limitations in terms of speed and accuracy in different situations. Viola-Jones and KLT are two well-known algorithms used for real time face detection and they work differently. Viola-Jones algorithm is much faster even though it is proved to be erroneous in some specific circumstances. KLT is basically used for tracking by points feature in frames. In this paper, we have given an overview of these two algorithms and made comparative analysis amongst them. Later we have proposed a method that improves the accuracy of Viola-Jones algorithm by removing/minimizing false detection.

Keywords: Face Detection, Viola-Jones algorithm, KLT algorithm, Error Correction, Individual detection.


By | 2017-07-06T07:27:46+00:00 July 6th, 2017|General, Publications|0 Comments


Shrikant malge1, Snehal D. Chaudhary2, Priyanka Paygude3

2,3Assistant Professor, Department of Information Technology,

Bharati Vidyapeeth College of Engineering, Pune, India


These days, we survive in large interrelated globe which produces a huge information every day, starting from the logging files of the users of social networks, search engines, e-mail clients to machine generated data as from the real-time monitoring of sensor networks for dams or bridges, and various vehicles such as airplanes, cars or ships. It also comprises private, skilled, communal data and so on. Artificial intelligent as well as inter connection direct to an unpredicted expansion of information. The Web is an input retriever for an information development. The web produced information previously go beyond the presented storage space. As 2011 concern in a region recognized as a big data that has been enlarged aggressively. Since the rising exploit of a big data as well as increasing capacity of big data, therefore protection of big data is measured critical. This complexity rotates into an unease to compact when believing that the data id uploading on to the cloud storage. In this paper we have proposed two different symmetric encryption algorithms such as AES and SERPENT to handle the big data on the cloud storage space with safe data access. Also we have proposed a new policy update method which helps to capable data access control in the dynamic policy update for big data in the cloud storage. In this, same input key is utilized for the both encryption and decryption operation. Here two types of files are handling as an input such as Text file and image file for encryption and decryption. In experimental result, result of both algorithms is shown with the help of graphs with two different parameters such as Time and Number of files. We have also shown the comparative analysis of both algorithms that shows AES encryption algorithm gives superior performance than the serpent encryption algorithm.

Keywords: Access control, protection, Encryption, Decryption, big Data, cloud storage.


By | 2017-07-06T07:23:53+00:00 July 6th, 2017|General, Publications|0 Comments


Mamatha H.K. 1, Chaithra N2, S Balasubramanian 3

 1Department of Health System Management Studies, Jagadguru Sri Shivarathreeshwara University, Mysuru, India

 2Faculty of Life Sciences, Jagadguru Sri Shivarathreeshwara University, Mysuru, India

3Jagadguru Sri Shivarathreeshwara University, Mysuru, India


Performance evaluation of faculty in higher education institutions is a challenge for education managers considering the diversity of disciplines. Though University Grants Commission (UGC) has given a format based on Academic Performance Indicators (API) many a times education managers face uncertainty in making decisions on the quality of performance as the scores are numerical and do not consider the subjective component involved in teaching learning process. The API format also includes several criteria like teaching, co curricular, research and feedback. The score of faculty across all these criteria indicates their overall performance which is used in decision making with regard to increment, promotion, etc. Measuring performance without considering the variation in each of these criteria may not yield good decisions. Hence, a statistical model which can assess the dependency and independency of the criteria was used to evaluate which of these criteria have maximum influence on the overall performance scores of the faculty.
In this study, multiple linear regression model is used to assess faculty performance wherein the total score of the faculty API score is taken as dependent variable and the teaching learning evaluation, co curricular activities, research and feedback are considered as independent variables. Factor scores obtained by factor analysis were used for multiple linear regression analysis using the SPSS software.

Key words: Faculty performance evaluation, higher education institutions, multiple linear regression analysis.


By | 2017-07-06T07:15:13+00:00 July 6th, 2017|General, Publications|0 Comments


Ayush Agarwal 1, Rishabh Bajaj 1, Rachit Sharma 1 , Shubham Dubey 1 , Ms. Shrikala Deshmukh 1 , Ms Snehal Chaudhary 1 , Ms Priyanka Paygude 1

 1Department of Information Technology Bharati Vidyapeeth Deemed University, College Of Engineering, Pune, India



 Voting is the most important thing which makes India the most democratic country. Voting is done to elect a candidate and make him/her the leader of a specific area which can be even large as a country. People have faith in that candidate, they hope that he will work in the favour of the society. The elected candidate has many powers which are provided to him by the government. These power can be misused too. There are many people who want to be elected for that power. In today’s world security and authenticity is very important aspect and people want that only the most deserving candidate should win. Every single vote is important. All the votes given by the public should be kept private and its authenticity should be maintained throughout the election process. We currently have a voting system which makes a person to physically go to an allotted booth which is given by the election commission of India according to the area in which the person is residing. There is no such procedure which allows us to vote by sitting in our homes. This is the era of internet, if we are having such facility then why not use it. The main problem in this procedure is security and authenticity. This particular paper proposes online voting application that can be operated remotely from your computers, smartphones, tablets, in short from every device which can run internet. By using this application user can vote online. The application provides verification through AADHAAR card for security and authenticity. The principle aim of this paper is actually providing an online voting application to vote and elect the most deserving candidate for a specific area. This application is quite secure and votes are authentic as it is using AADHAAR card for verification.

 Keywords: Voting, AADHAAR card registration, AADHAAR card verification, Election Commission of India.


By | 2017-07-04T11:02:10+00:00 July 4th, 2017|General, Publications|0 Comments


Jindal Rajni 1, Singla Abhinash 2

 1 Student, Mtech CSE Department, Bhai Gurdas Institute of Engineering &  Technology, Sangrur

 2 Head  CSE Department, Bhai Gurdas Institute of Engineering &  Technology, Sangrur


The basic objective of this paper is provide an algorithm to solve the problem, when all clients are relying on a single server. Due to planned server downtime or from server failures the database becomes unavailable; all of the remote workers will be disconnected from their data. Data is stored on their system (user system). When the user connected to the internet data automatically sink from their client system to the server in serial order and it also works on file handling. When the system is disconnected from the network all the files (images) uploaded by user, saved on the client machine folder when it is again connected to the server, automatically files (images) transferred from client to server.

Keywords: Plugged off mode, Plugged In mode, Web Functions, Web Services, Data Synchronization


By | 2017-07-04T10:56:13+00:00 July 4th, 2017|General, Publications|0 Comments


Onintra Poobrasert 1, Thaphat Mupattararot 2, Lattapol Sae-aue3

1,2,3  National Electronics and Computer Technology Center,  National Science and Technology Development Agency, Thailand


This study represents an experiment on the efficiency of an assistive technology tool for learning disabilities in writing. The tool used in this study is Thai Word Prediction add-in for Microsoft Word. The hypothesis was to gain results of typing using the add-in better than results of writing without using the add-in. A representative sample included 8 Thai students with learning disabilities in writing from grade 4 to 6. The pre-test and post-test are presented in this study. Moreover, the authors conducted this study with 3 main methods. The first method is to write 2 stories from dictation without using Thai Word Prediction add-in and then compare the results to the typing with the same stories using the add-in. The second method is to compose a story for each 2 given topics by writing without using the add-in and then compare the results to the typing using the add-in. The last method was to compose a story for each 2 given pictures by writing without using the add-in and then typing using the add-in. Results for the first method shows that the scores of correct words from typing using Thai Word Prediction add-in are between 80-100%. Hence, the scores of correct words from writing without using the add-in are between 32-91%. Results for the second method shows that the scores of correct words from typing using the add-in is greater than the scores of correct words from writing without using the add-in for 6 out of 8 students. For the third method, there are 7 out of 8 students have the correct words of typing using the add-in better than writing without using the add-in. Additionally results show that all students could compose stories from given orders longer and more correct using the add-in than the writing without add-in.

Keywords: Learning disabilities, Microsoft word, Word completion, Word prediction, Writing disorder, Single subject research design



By | 2017-07-04T10:50:03+00:00 July 4th, 2017|General, Publications|0 Comments


Manoj Manjunath1, Mohith.D2, N.V.Swati3, Poornima Sridhar4, Nagarathna .N5

1,2,3,4 Students, 5 Assoc.Professor, Department of Computer Science & Engineering,

B.M.S College of Engineering, Bengaluru, India

The paper gives an account of different image deduplication techniques employed at the client side and the server side. Deduplication is the process of removal of duplicate data. This technique can be used for both text and image data. This technique considerably reduces the data storage and also overhead of huge data transmission. Apart from the conventional methods of deduplication, the use of various perceptual hashing algorithms, their strengths and weaknesses are studied. Perceptual hashing will be used as hashing algorithm in Content Aware Image deduplication. It finds identical images and also differentiates dissimilar images efficiently.

Keywords: Data storage, Data transmission, Deduplication, Content-aware image deduplication, Perceptual hashing.


By | 2017-07-04T10:45:20+00:00 July 4th, 2017|General, Publications|0 Comments


Dhanashree H. Parab, Akshay Nambly, Shraddhali N. Patil and Linda Mary John

Department of Computer Engineering Mumbai University, India



The Cryptography is a practice used in hiding the confidential information into a form in which the hidden information is accessed only by the intended person. Traditionally, there are many ways which can be used like Additive cipher, Substitution cipher, Caeser Cipher etc. for encryption and decryption of information. The earlier techniques referred as Ciphers, which are a set of steps (algorithm) for encryption and the corresponding description. The basic elements used were the alphabets. Here in our work, we are looking after hiding the information using musical notes and their frequencies so that they cannot be recognized which is known as Musical Cryptography. The paper focuses on developing a system which encrypts the information into a music and then decrypts it into its text format using the Genetic Algorithm. We create a population, find fitness and perform crossover to get the best output.

Keywords: Cryptography, Musical Cryptography, Ciphers, Encryption, Decryption, Genetic Algorithm, Population, Fitness, Mutation


By | 2017-07-04T10:41:18+00:00 July 4th, 2017|General, Publications|0 Comments


Shweta Malwatkar 1, Rekha Sugandhi2, Dr. Anjali R. Mahajan 3

 1Department of Computer Engineering MIT College of Engineering Pune, India

2Department of Computer Engineering, MIT College of Engineering Pune, India

3Department of computer and IT, Govt. Polytechnic, Nagpur, India


The expressions on the face is the key factor of communication among human beings. Facial expression recognition is the computer based automatic process. This process works in four major parts such as: Face detection in input image or video, preprocessing of the extracted face region, facial expression feature extraction and classification. Classification of facial expression is based on either prototype facial expression such as happy, sad, anger etc. or on Facial Action units. This paper concentrates on Facial Action Coding System (FACS) for classification. Initially system takes the video as an input and divide it into frames by using Xuggler Library. These frames are then preprocessed with Haar like features, for face, eyes and mouth detection. After this, facial point localization and facial feature point are required in FACS based classification. For improving the classification accuracy and speed and to reduce the emotion confusion, Concept Map is used. Experimental results prove that the FACS and Concept map based classification will improve the accuracy of classification results.

Keywords: Affect recognition, feature extraction, face detection, concept map.



By | 2017-07-01T10:15:09+00:00 July 1st, 2017|Conference Publications|0 Comments

Enhancing Tagging of Digital Documents Using OCR

Mahendra K. Ugale1, Madhuri S.Joshi2

[1][2] Dept of Computer Science and Engineering, Jawaharlal Nehru Engineering College, Aurangabad, M.S., India


A document management system is based on management of digital documenID_62 Enhancing Tagging of Digital Documents Using OCRts used to track search and store documents and reduce paper. This is done by converting documents and other papers into digital form like pdf i.e. Portable File Format by using OCR. Document image analysis is important aspect in automating the offices. In most of the cases office consists numerous documents presenting a mixture of handwritten and printed text. The recognition of such documents is a challenging task for OCR designers. In this paper comparison is presented of different tools for searching handwritten and printed text from document images by tagging documents automatically and manually. Hence classification of handwritten and printed text from document image is important to optimize the OCR accuracy.

Keywords: OCR, Tagging, Handwritten and Printed Text, Document Management System.


By | 2017-07-01T10:11:09+00:00 July 1st, 2017|Conference Publications|0 Comments

Fire Detection Using Image Processing Using Raspberry PI

Anita Thengade1, Prateek Mishra 2,Raunak Kshatriya3,Rahul Mhaskar4,Pankaj Bodhe5

1Department of Computer Engineering MITCOE Pune,India

2,3,4,5 Department of Computer Engineering MITCOE Pune,India


The fire can be recognised effectively through video image processing. Traditional fire recognition system based on video image analysis do not analyze the color,texture and low level visual feature information of specific fire image.The validation performance is not good,the fire image scene recognition method is proposed on image processing .A contour is a curve joining all the continuous points along the boundary with the same color value. The detecting of contours in an image is very useful if you want to detect the boundaries in images.Gray processing and binaryzation process are taken into consideration for fire image.The normalization process is taken for filtering background interface.The refined video digital image is analyzed . Fire Sensor is used to collect the visual feature of fire video monitoring images.Simulation shows algorithm has good efficiency it can recognize campus fire accurately and gives real time alarming.

Keywords: Pi,Diode,Resistor,Image Processing,Binaryzation


By | 2017-07-01T10:07:18+00:00 July 1st, 2017|Conference Publications|0 Comments


Akshay Khedaskar 1, Suyog Gaikwad 2 , Ruchir Tawade3 , Megha Naik4 

1Department of Computer Engineering, Pillai HOC College of Engineering and Technology, Rasayani. Raigad, India akhedaskar@gmail.com

2Department of Computer Engineering, Pillai HOC College of Engineering and Technology, Rasayani. Raigad, India akhedaskar@gmail.com

3Department of Computer Engineering, Pillai HOC College of Engineering and Technology, Rasayani. Raigad, India, rtruchir33@gmail.com

4 Department of Computer Engineering, Pillai HOC College of Engineering and Technology, Rasayani. Raigad, India, meghanaik09@gmail.com 


This is an Android application which is very useful for college students. In the previous system, all the information must be viewed in a website. At that time searching any information is too difficult and it takes a lot of time to search the particular website for each and everything like (Question papers, Aptitude Test, Information of Syllabus etc.), and it needs additional browser to search any website. Hence, to overcome this problem a smart phone based application using Android can be used to make this process easier and faster. Android is an open source Linux based system developed by Google. We will be using it to accomplish our daily task. One application that falls into this category is the Android Application for College. This College application provides a wide range of useful information which splits into several functionalities. These include: Syllabus, Question papers, Aptitude Questions, College Notices or Notifications, What’s New Today and Important Videos. Users can install this application in their android mobile to view all this Information and use it.
The design and implementation of this system is to provide services in institute. This application is used to provide comprehensive student information and user interface. The system plans for student user interface, allowing students to access resources as provided by college. Previously, college students relied heavily on browsers which had its own disadvantages like searching all websites for each and everything, but this system helps us to maintain all required information in single app.

1.2 This application decreases time needed to search all information. This system provides a simple interface for the maintenance of student information. This app will provide information in an easy and interactive manner which provides facilities like online registration, questionnaire system, syllabus viewing, providing Aptitude Questions, Important Educational videos related to Syllabus, College Notification and many more activities. The results can also be displayed on web server.

Keywords: Android, Cloud Storage, Library Books, Notification, Syllabus, Question Paper, Aptitude Questions, Important Videos.


By | 2017-07-01T10:03:35+00:00 July 1st, 2017|Conference Publications|0 Comments


Akshay Sharma1, Ameya Ambekar2, Pavan Dahale3, Priyansh Bendre4, Prasad Halgaonkar5

1,2,3,4,5Computer Engineering Department MIT College of Engineering, Pune India


NFC holds the potential of eliminating the requirement of setting up high-cost and complex IT infrastructure for solutions and services. It provides flexible and secure environment for easy access and development through lightweight portable devices. In NFC, a small amount of information stored in the tag is used to authenticate a device. The goal of this paper is twofold. First in this paper we trying to compare various wireless data transfer technology standards. And Secondly this paper further gives details about the benefits of using NFC over other mediums for secure authentication.

Keywords: NFC, Bluetooth, BLE, Infrared, RFID



By | 2017-07-01T09:58:38+00:00 July 1st, 2017|Conference Publications|0 Comments


Sneha Khandare1, Ifreen Raveen2, Sachin Gahlot3, Saurabh Kumar4, Mr. Pranav Shriram5  

1Department of Information Technology MIT Academy of Engineering, Pune, India


Smartphone based systems have gained popularity over the years and android is about to become the most widely used OS on mobile phones. In the rapidly advancing world of technology, mobile applications are a continuously growing segment of the global mobile market. Mobile applications are evolving at an astounding pace to give users a rich and fast user experience. The intention is to provide surveillance of an area with motion detection feature remotely through an android application. The CCTV captures images and videos and sends them to the server, which operates and handles motion detectors and video cameras for remote surveillance, streams live video and records it for future playback. The server transmits this data to the authorized smart phones through HTTP connections where we can access them through an application. There are two important aspects, first is the image processing. Second is the connectivity between the server and the CCTV camera and server and the application.

Keywords: Surveillance, Android Application, Intrusion Detection, Image Processing, Remote Monitoring



By | 2017-07-01T09:54:32+00:00 July 1st, 2017|Conference Publications|0 Comments


H. K. Khanuja, Amruta Mujumdar, Manashree Waghmare, Mrudula Kulkarni, Mrunal Bajaj

Department of Computer Engineering Marathwada Mitra Mandal’s College of Engineering Pune, India


Majority of the digital libraries implemented in worldwide use server-client architecture. Such libraries require a large infrastructure investment in server setups and maintenance. Servers are also very difficult to scale as they are based on changing institutional requirements and needs. These drawbacks can be overcome by using a distributed database system. Apache Hadoop is one of the large scale data processing projects that support data-intensive distributed applications. Hadoop applications utilize a distributed file system for data storage called Hadoop Distributed File System (HDFS). Apache Solr is used for processing and searching the required data. Apache Solr provides search features like Full-Text Searching, Hit Highlighting. Solr searches the index instead of searching the text. Because of this feature Solr achieve fast search responses.

Keywords— Hadoop, HDFS, Digital Library, Replication, SOLR




By | 2017-07-01T09:51:02+00:00 July 1st, 2017|Conference Publications|0 Comments

Topic Modeling on Online News and News Clustering

Aashka Sahni1 and Prof. Sushila Palwe2

Department of Computer Engineering, MIT College of Engineering, Pune 


News media includes print media, broadcast news and internet. Print media contains newspapers, news magazines, broadcast news contains radio and television, while internet contains online newspapers, news blogs, etc. The online news has been the prevalent form of information on the internet. Often, the occurrence of the same event or happening is depicted differently in different news websites or sources due to the varied perceptions of the same circumstance. Proposed system intends to collect news data from such diverse sources, capture the varied perceptions, summarize and present them at one place. Another goal of the proposed system includes detecting topics accurately in case of short news data. Previous approaches like LDA and its variants are able to identify topics efficiently for long texts (news), however, fail to do so in the case of short texts (news) due to data sparsity problem. Since sophisticated signals are delivered by the short news, it is an importnat resource for topic modeling, however, the issues of acute sparsity and irregularity are prevalent. These pose new difficulties to existing topic models, like LDA and its variations. In this paper, a lucid but generic explanation for topic modeling in online news has been provided. System presents a word co-occurrence network based model named WNTM, which works for both long as well as short news articles by managing the sparsity and imbalance issues simultaneously. WNTM is modeled by assigning and reassigning (according to probability calculation) a topic to every word in the document rather than modeling topics for every document. It effectively improves the density of information space without wasting much time and space complexity. Along these lines, the rich context saved in the word-word space likewise ensures to detect new and uncommon topics with convincing quality. The system extracts real time online news data and uses this data for system implementation. Firstly, topic modeling algorithm is applied on this online news data to identify the key topic of the incoming news and also to identify the most trending topic. Once we identify the topic of news, the system uses k-means document clustering algorithm to cluster all latest news associated to a particular topic together. Likewise, classify the news on the basis of topic. After clustering, generation of the summary is done from the output and we intend to present the summarized news along with the topic to the user.

Keywords: Data mining, Topic modeling, document clustering, online news



By | 2017-07-01T09:45:07+00:00 July 1st, 2017|Conference Publications|0 Comments


Swapnil Vaidya1, Jayshree Aher2

1,2Department of Computer Engineering MITCOE, Pune, India


Data mining also known as knowledge discovery, is a process of discovering useful information or patterns from different perspectives. Data mining involves the intersection of other domain fields, such as artificial intelligence, machine learning, statistics etc. Text mining is a subset of data mining, as data can be of any form, text, image, video, signals from sensor devices etc. Text data includes text files, scholarly articles, short text (tweets), may be tabular representation as in relational databases etc. Documents are in natural language, i.e. English (most preferably), which needs to be processed in more efficient way. Natural Language Processing is a method which can increase the efficiency of text mining, by converting the text documents into more understandable form. In natural languages, most important task is preprocessing of text, we need the data in correct format for analyzing it efficiently. To extract some information from a bulk of data (unstructured text data), which will lead to some meaningful pattern or knowledge preprocessing plays very important role. This paper presents some preprocessing methods which are involved in preprocessing of unstructured text data, such as social networks (twitter, facebook, etc.) applications of Natural Language Processing.

Keywords: ID_50_SwapnilVaidya_NLP (1) (1)



By | 2017-07-01T09:38:20+00:00 July 1st, 2017|Conference Publications|0 Comments


Ashish Varghese 1, Tejaswini Kasture 2, Yash Gupta 3, Prithviraj Patil  4

 Department of Computer Engineering MIT College of Engineering, India 


Autism is a lifelong development disability which affects the overall understanding of the autistic child and causes problems with speech and nonverbal communication. Autistic children have to be taught again and again which makes it a tedious task for the parents and the teachers to teach them. The child’s ability to function can be improved by the help of treatments and services. These treatments and services can be provided using the latest trends in mobile technology. So for this reason we have proposed a system which eliminates the repetitive efforts required by teachers and parents by providing a customizable, adaptable and cost efficient system. This system is built using open source tools like FFmpeg and Opensmile. It adapts itself according to the behavior and learning of the child and provides the next level of learning depending on the learning level of the child.

Keywords: ID_49_An android based assisting and monitoring tool for autistic children (1)


By | 2017-07-01T09:34:33+00:00 July 1st, 2017|Conference Publications|0 Comments


Ankita Kale1, Jayshree Aher2  

1Department of Computer Engineering MIT College of Engineering Pune, India  

2Department of Computer Engineering, MIT College of Engineering Pune, India 


We have number of reviews every day on the internet about different products, movies and what not, consist of valuable information like reviewer’s emotion, opinion or attitude. Capturing those sentiments of customers from that informal text is task of sentiment Analysis. This paper outlines briefly an approach based on Word Alignment Model for deriving sentiments of the reviewer at deeper level. With the Electronic Product Review dataset, system preprocess the data to obtain the segments for the Word Alignment Model which is trained using EM algorithm in unsupervised manner. Partial links from the syntactic parser is incorporated to improve the result. Association Relation between extracted words are calculated along with the confidence value for each word individually. With these probabilistic values, a graph called Opinion Association Graph depicts the relativeness between relevant product features and sentiments.

Keywords: ID_48_Ankita_Kale_CRC



By | 2017-07-01T09:31:54+00:00 July 1st, 2017|Conference Publications|0 Comments


Pooja Bhatia, Pratik P. Watwani

1,2Department of Computer Engineering, Vivekanand Education Society’s Institute of Technology, Mumbai, India


Sentiment Analysis(SA) concentrates on investigating sentiments of regular dialect by the methods of using different algorithms and procedures. Sentiments, Opinions, Views all stand up to be the mainstays of basic decision making in today’s time. Each Product, service gets a supposition or estimation from its users. Organizations expect their products and services to beat the class. In this, SA comes into the picture. Even though SA has a great many application, there has been a constrained measure of research that has been done. Web permits the sharing of data, there is a tremendous measure of unstructured information that shapes the sentiments, feelings, considerations and so forth. Sentiment Analysis is vital, because it helps in deciding the extremity of sentiments of users. This paper focuses on determining how SA can be used by an organization or a service provider to comprehend their customers feedback and sentiments and along these lines make important steps as required.

Keywords: ID_46_[CAMERA READY]Evaluating Product Service Review via Sentiment Analysis



By | 2017-07-01T09:28:31+00:00 July 1st, 2017|Conference Publications|0 Comments