Sujata Khedkar | Mumbai - Academia.edu (original) (raw)

Papers by Sujata Khedkar

International Journal for Research in Applied Science and Engineering Technology, Apr 30, 2018

Every person is unique", we have been hearing this since ages. Every person has a unique identity... more Every person is unique", we have been hearing this since ages. Every person has a unique identity, a unique fingerprint, a unique retina and a lot more. These features play a vital role in identification of individuals for security purposes. Unfortunately, when it comes to security of written pieces or words from an individual, these primary unique identities are futile. One cannot identify a writer from a written piece of text on the basis of retina or fingerprint scans, sometimes even the signature can be forged, in such situations for security purposes and intellectual property rights it becomes very important to identify the true author. Stylometry plays an important role in this. Every author has a unique style of writing, measure of this style of writing is called Stylometry. This paper proposes to identify authors from text based on their style of writing. First a data set consisting of articles, short stories and emails will be used to train the system for multiple authors, then a random text would be given to the system to identify the author correctly, if the author predicted by the system is similar to the author claimed then the information is authentic otherwise the author claiming to be the writer is a fraud. For stylometry, over the ages, many features have been focused on, but this paper proposes new features to be used for this purpose. While writing, there are many unconscious styles that are incorporated by the author, these features have been unnoticed till date, but can play a vital role in accurate and fast identification of authors. These features include: 'intellectual property right', 'chapter length', 'the importance of a word with respect to the other words in a document' and frequency of particular words per thousand words. The algorithms used to train the system can be Decision tree, Naive Bayesian or Multilayer Perceptron.

In today's world, the daily hustle-bustle does not permit a human being to devote time for ma... more In today's world, the daily hustle-bustle does not permit a human being to devote time for manually summarizing various lengthy documents. Hence it is of utmost importance to devise an application that will facilitate automated text summarization. Not only will this application save time but also render higher scope of efficiency. This application will allow the user to automatically summarize relevant information from various sources.

Social Science Research Network, 2019

Apple Academic Press eBooks, May 23, 2022

The project revolves around the idea of scene understanding purpose based on the video input, thu... more The project revolves around the idea of scene understanding purpose based on the video input, thus not continuously monitoring the feed manually. The videos are extracted into the form of raw video frames and using 2D-3D CNN, the feature vector is extracted. Using You Only Look Once - version 3 (YOLOv3) algorithm, the objects present in a particular frame is identified. Also, the count of the objects is stored. The pose of people present in the frames is estimated for identification of movements. Through this, the actions are recognized as being performed by the people. All the words that are formed through the above three methods count to input to the LSTM cell. This cell selects the words based on their probabilities and confidence rate and forms a natural language sentence for the user to understand. Finally, the generated output can be modified or changed completely by the user using Human-in-the-loop concept, if required. The machine will retrain itself based on this input and generate better results next time. The central model is capable of identifying as well as discriminating between types of elements which are required for this project. This project was built as a continuation of the previous system, which works on object identification from live video input from drones. In the case of poor network issues, when sending video data becomes difficult, the data is sent in textual format.

Lecture Notes in Networks and Systems, 2018

International Journal for Research in Applied Science and Engineering Technology

Text preprocessing is the most essential and foremost step for any Machine Learning model. The ra... more Text preprocessing is the most essential and foremost step for any Machine Learning model. The raw data needs to be cleaned and pre-processed to get better performance. It is the method to clean the data and makes it ready to feed the data to the model. Text classification is the heart of many software systems that involve text documents processing. The purpose of text classification is to classify the text documents automatically into two or many defined categories. In this paper ,various preprocessing and classification approaches are used such as NLP, Machine Learning, etc from patent documents.

Algorithms for intelligent systems, 2023

2022 International Conference on Data Science, Agents & Artificial Intelligence (ICDSAAI)

Apple Academic Press eBooks, May 23, 2022

Proceedings of The International Conference on Emerging Trends in Artificial Intelligence and Smart Systems, THEETAS 2022, 16-17 April 2022, Jabalpur, India

Journal of emerging technologies and innovative research, Oct 1, 2021

The Objective of our assignment is to hold Social Distance amongst humans and to test Face Mask o... more The Objective of our assignment is to hold Social Distance amongst humans and to test Face Mask on face of peoples withinside the time of COVID-19. This version may be detecting actual time Face masks and Social Distancing which may be very vital on this pandemic situation. Since all of the schools, schools and places of work which are closed now will reopen quickly and they may be wanting a few generations to be secure and healthy.

ICT for Competitive Strategies, 2020

Advances in Intelligent Systems and Computing, 2019

Online reviews are very important in the customer’s decision-making process in selecting the appr... more Online reviews are very important in the customer’s decision-making process in selecting the appropriate products in the online shopping portal. These reviews are then analyzed by business organizations to understand customer sentiment w.r.t. product/service. Traditional sentiment analysis techniques identify only positive, negative or neutral sentiment w.r.t. reviews and does not consider informativeness of reviews while analyzing sentiment. The extreme opinions like Praise and complaint sentences are considered as a subset of positive and negative sentences and becomes difficult to find. Praise sentences are more descriptive in nature. Praises contain more nouns, adjectives, intensifiers as compared to plain positive sentences and complaint sentences contain more connectives and adverbs rather than the plain negative sentences. This paper proposes a Linguistic feature-based approach for review sentences filtering and Hybrid feature selection method for classifying review sentence as Praise or Complaint.

The aim of this system is to explore the influence of applying a game-based learning approach to ... more The aim of this system is to explore the influence of applying a game-based learning approach to nutrition education so that students could start learning with great enthusiasm and interest along with students, participants other age group can take advantage of this system i.e. to check their knowledge and to conduct a systematic literature review on empirical studies of how technologies influence young children's learning.

2020 International Conference on Electronics and Sustainable Communication Systems (ICESC), 2020

The project revolves around the idea of scene understanding purpose based on the video input, thu... more The project revolves around the idea of scene understanding purpose based on the video input, thus not continuously monitoring the feed manually. The videos are extracted into the form of raw video frames and using 2D-3D CNN, the feature vector is extracted. Using You Only Look Once - version 3 (YOLOv3) algorithm, the objects present in a particular frame is identified. Also, the count of the objects is stored. The pose of people present in the frames is estimated for identification of movements. Through this, the actions are recognized as being performed by the people. All the words that are formed through the above three methods count to input to the LSTM cell. This cell selects the words based on their probabilities and confidence rate and forms a natural language sentence for the user to understand. Finally, the generated output can be modified or changed completely by the user using Human-in-the-loop concept, if required. The machine will retrain itself based on this input and generate better results next time. The central model is capable of identifying as well as discriminating between types of elements which are required for this project. This project was built as a continuation of the previous system, which works on object identification from live video input from drones. In the case of poor network issues, when sending video data becomes difficult, the data is sent in textual format.

2018 International Conference on Smart City and Emerging Technology (ICSCET), 2018

“Every person is unique”, we have been hearing this since ages. Every person has a unique identit... more “Every person is unique”, we have been hearing this since ages. Every person has a unique identity, a unique fingerprint, a unique retina and a lot more. These features playa vital role in identification of individuals for security purposes. Unfortunately, when it comes to security of written pieces or words from an individual, these primary unique identities are futile. One cannot identify a writer from a written piece of text on the basis of retina or fingerprint scans, sometimes even the signature can be forged, in such situations for security purposes and intellectual property rights it becomes very important to identify the true author. Stylometry plays an important role in this. Every author has a unique style of writing, measure of this style of writing is called Stylometry. This paper proposes to identify authors from text based on their style of writing. First a data set consisting of articles, short stories and emails will be used to train the system for multiple authors, then a random text would be given to the system to identify the author correctly, if the author predicted by the system is similar to the author claimed then the information is authentic otherwise the author claiming to be the writer is a fraud. For stylometry, over the ages, many features have been focused on, but this paper proposes new features to be used for this purpose. While writing, there are many unconscious styles that are incorporated by the author, these features have been unnoticed till date, but can playa vital role in accurate and fast identification of authors. These features include: ‘intellectual property right’, ‘chapter length’ and frequency of particular words per thousand words. The algorithms used to train the system can be Decision tree, Naive Bayesian or Multilayer Perceptron.

With the increasing popularity of development of chatbots to overcome restrictions of complex use... more With the increasing popularity of development of chatbots to overcome restrictions of complex user interface design and inconvenience to the user in requesting data and services, we felt that there is a strong need for an analytics platform for chatbot developers through which they can implement analytics on the information collected by the chatbot to better understand the needs of the user. The paper tries to identify and find solutions for two types of analytics needed by the chatbot. One is to summarize the reviews and nature of a product or service from multiple sources on the internet. Other caters to capture and analyse user engagement characteristics to provide better recommendations.

International Journal of Advanced Research in Computer Science, 2017

A taxi fleet management system is presented .The system consists of Clustering , Neuro fuzzy syst... more A taxi fleet management system is presented .The system consists of Clustering , Neuro fuzzy systems and Particle Swarm Optimization methodologies. The proposed system aims at maximizing revenue of cabs as individual entities and the cab aggregator simultaneously. Clustering of pick up requests is carried out using a variant of DBSCAN which uses Delaunay triangulation to recognise fare hotspots. Neuro Fuzzy system is used to evaluate the eligibility of taxis to contest for these hotspots .The Neuro Fuzzy System is trained using Particle Swarm Optimization method. Intelligent swarming of taxis according to their eligibilities for the hotspots is performed to maximize revenue of both cab aggregators and cabs. Keywords: PSO, TSK Model, Taxi Fleet Management, Neuro Fuzzy Systems, Clustering, Fleet Management , Particle Swarm Optimization , Swarm Intelligence.

International Journal for Research in Applied Science and Engineering Technology, Apr 30, 2018

Every person is unique", we have been hearing this since ages. Every person has a unique identity... more Every person is unique", we have been hearing this since ages. Every person has a unique identity, a unique fingerprint, a unique retina and a lot more. These features play a vital role in identification of individuals for security purposes. Unfortunately, when it comes to security of written pieces or words from an individual, these primary unique identities are futile. One cannot identify a writer from a written piece of text on the basis of retina or fingerprint scans, sometimes even the signature can be forged, in such situations for security purposes and intellectual property rights it becomes very important to identify the true author. Stylometry plays an important role in this. Every author has a unique style of writing, measure of this style of writing is called Stylometry. This paper proposes to identify authors from text based on their style of writing. First a data set consisting of articles, short stories and emails will be used to train the system for multiple authors, then a random text would be given to the system to identify the author correctly, if the author predicted by the system is similar to the author claimed then the information is authentic otherwise the author claiming to be the writer is a fraud. For stylometry, over the ages, many features have been focused on, but this paper proposes new features to be used for this purpose. While writing, there are many unconscious styles that are incorporated by the author, these features have been unnoticed till date, but can play a vital role in accurate and fast identification of authors. These features include: 'intellectual property right', 'chapter length', 'the importance of a word with respect to the other words in a document' and frequency of particular words per thousand words. The algorithms used to train the system can be Decision tree, Naive Bayesian or Multilayer Perceptron.

In today's world, the daily hustle-bustle does not permit a human being to devote time for ma... more In today's world, the daily hustle-bustle does not permit a human being to devote time for manually summarizing various lengthy documents. Hence it is of utmost importance to devise an application that will facilitate automated text summarization. Not only will this application save time but also render higher scope of efficiency. This application will allow the user to automatically summarize relevant information from various sources.

Social Science Research Network, 2019

Apple Academic Press eBooks, May 23, 2022

The project revolves around the idea of scene understanding purpose based on the video input, thu... more The project revolves around the idea of scene understanding purpose based on the video input, thus not continuously monitoring the feed manually. The videos are extracted into the form of raw video frames and using 2D-3D CNN, the feature vector is extracted. Using You Only Look Once - version 3 (YOLOv3) algorithm, the objects present in a particular frame is identified. Also, the count of the objects is stored. The pose of people present in the frames is estimated for identification of movements. Through this, the actions are recognized as being performed by the people. All the words that are formed through the above three methods count to input to the LSTM cell. This cell selects the words based on their probabilities and confidence rate and forms a natural language sentence for the user to understand. Finally, the generated output can be modified or changed completely by the user using Human-in-the-loop concept, if required. The machine will retrain itself based on this input and generate better results next time. The central model is capable of identifying as well as discriminating between types of elements which are required for this project. This project was built as a continuation of the previous system, which works on object identification from live video input from drones. In the case of poor network issues, when sending video data becomes difficult, the data is sent in textual format.

Lecture Notes in Networks and Systems, 2018

International Journal for Research in Applied Science and Engineering Technology

Text preprocessing is the most essential and foremost step for any Machine Learning model. The ra... more Text preprocessing is the most essential and foremost step for any Machine Learning model. The raw data needs to be cleaned and pre-processed to get better performance. It is the method to clean the data and makes it ready to feed the data to the model. Text classification is the heart of many software systems that involve text documents processing. The purpose of text classification is to classify the text documents automatically into two or many defined categories. In this paper ,various preprocessing and classification approaches are used such as NLP, Machine Learning, etc from patent documents.

Algorithms for intelligent systems, 2023

2022 International Conference on Data Science, Agents & Artificial Intelligence (ICDSAAI)

Apple Academic Press eBooks, May 23, 2022

Proceedings of The International Conference on Emerging Trends in Artificial Intelligence and Smart Systems, THEETAS 2022, 16-17 April 2022, Jabalpur, India

Journal of emerging technologies and innovative research, Oct 1, 2021

The Objective of our assignment is to hold Social Distance amongst humans and to test Face Mask o... more The Objective of our assignment is to hold Social Distance amongst humans and to test Face Mask on face of peoples withinside the time of COVID-19. This version may be detecting actual time Face masks and Social Distancing which may be very vital on this pandemic situation. Since all of the schools, schools and places of work which are closed now will reopen quickly and they may be wanting a few generations to be secure and healthy.

ICT for Competitive Strategies, 2020

Advances in Intelligent Systems and Computing, 2019

Online reviews are very important in the customer’s decision-making process in selecting the appr... more Online reviews are very important in the customer’s decision-making process in selecting the appropriate products in the online shopping portal. These reviews are then analyzed by business organizations to understand customer sentiment w.r.t. product/service. Traditional sentiment analysis techniques identify only positive, negative or neutral sentiment w.r.t. reviews and does not consider informativeness of reviews while analyzing sentiment. The extreme opinions like Praise and complaint sentences are considered as a subset of positive and negative sentences and becomes difficult to find. Praise sentences are more descriptive in nature. Praises contain more nouns, adjectives, intensifiers as compared to plain positive sentences and complaint sentences contain more connectives and adverbs rather than the plain negative sentences. This paper proposes a Linguistic feature-based approach for review sentences filtering and Hybrid feature selection method for classifying review sentence as Praise or Complaint.

The aim of this system is to explore the influence of applying a game-based learning approach to ... more The aim of this system is to explore the influence of applying a game-based learning approach to nutrition education so that students could start learning with great enthusiasm and interest along with students, participants other age group can take advantage of this system i.e. to check their knowledge and to conduct a systematic literature review on empirical studies of how technologies influence young children's learning.

2020 International Conference on Electronics and Sustainable Communication Systems (ICESC), 2020

The project revolves around the idea of scene understanding purpose based on the video input, thu... more The project revolves around the idea of scene understanding purpose based on the video input, thus not continuously monitoring the feed manually. The videos are extracted into the form of raw video frames and using 2D-3D CNN, the feature vector is extracted. Using You Only Look Once - version 3 (YOLOv3) algorithm, the objects present in a particular frame is identified. Also, the count of the objects is stored. The pose of people present in the frames is estimated for identification of movements. Through this, the actions are recognized as being performed by the people. All the words that are formed through the above three methods count to input to the LSTM cell. This cell selects the words based on their probabilities and confidence rate and forms a natural language sentence for the user to understand. Finally, the generated output can be modified or changed completely by the user using Human-in-the-loop concept, if required. The machine will retrain itself based on this input and generate better results next time. The central model is capable of identifying as well as discriminating between types of elements which are required for this project. This project was built as a continuation of the previous system, which works on object identification from live video input from drones. In the case of poor network issues, when sending video data becomes difficult, the data is sent in textual format.

2018 International Conference on Smart City and Emerging Technology (ICSCET), 2018

“Every person is unique”, we have been hearing this since ages. Every person has a unique identit... more “Every person is unique”, we have been hearing this since ages. Every person has a unique identity, a unique fingerprint, a unique retina and a lot more. These features playa vital role in identification of individuals for security purposes. Unfortunately, when it comes to security of written pieces or words from an individual, these primary unique identities are futile. One cannot identify a writer from a written piece of text on the basis of retina or fingerprint scans, sometimes even the signature can be forged, in such situations for security purposes and intellectual property rights it becomes very important to identify the true author. Stylometry plays an important role in this. Every author has a unique style of writing, measure of this style of writing is called Stylometry. This paper proposes to identify authors from text based on their style of writing. First a data set consisting of articles, short stories and emails will be used to train the system for multiple authors, then a random text would be given to the system to identify the author correctly, if the author predicted by the system is similar to the author claimed then the information is authentic otherwise the author claiming to be the writer is a fraud. For stylometry, over the ages, many features have been focused on, but this paper proposes new features to be used for this purpose. While writing, there are many unconscious styles that are incorporated by the author, these features have been unnoticed till date, but can playa vital role in accurate and fast identification of authors. These features include: ‘intellectual property right’, ‘chapter length’ and frequency of particular words per thousand words. The algorithms used to train the system can be Decision tree, Naive Bayesian or Multilayer Perceptron.

With the increasing popularity of development of chatbots to overcome restrictions of complex use... more With the increasing popularity of development of chatbots to overcome restrictions of complex user interface design and inconvenience to the user in requesting data and services, we felt that there is a strong need for an analytics platform for chatbot developers through which they can implement analytics on the information collected by the chatbot to better understand the needs of the user. The paper tries to identify and find solutions for two types of analytics needed by the chatbot. One is to summarize the reviews and nature of a product or service from multiple sources on the internet. Other caters to capture and analyse user engagement characteristics to provide better recommendations.

International Journal of Advanced Research in Computer Science, 2017

A taxi fleet management system is presented .The system consists of Clustering , Neuro fuzzy syst... more A taxi fleet management system is presented .The system consists of Clustering , Neuro fuzzy systems and Particle Swarm Optimization methodologies. The proposed system aims at maximizing revenue of cabs as individual entities and the cab aggregator simultaneously. Clustering of pick up requests is carried out using a variant of DBSCAN which uses Delaunay triangulation to recognise fare hotspots. Neuro Fuzzy system is used to evaluate the eligibility of taxis to contest for these hotspots .The Neuro Fuzzy System is trained using Particle Swarm Optimization method. Intelligent swarming of taxis according to their eligibilities for the hotspots is performed to maximize revenue of both cab aggregators and cabs. Keywords: PSO, TSK Model, Taxi Fleet Management, Neuro Fuzzy Systems, Clustering, Fleet Management , Particle Swarm Optimization , Swarm Intelligence.