Vrushali Y. Kulkarni
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Vrushali Y. Kulkarni has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
Asmita Kashid
2 shared publicationsPradeep K Sinha
2 shared publicationsPrajakta Joglekar
Department of Computer Engineering, MIT-World Peace University, Pune, India
2 shared publicationsPranali K. Kosamkar
MIT-WPU,School of Computer Engineering & Technology,Pune,India
2 shared publicationsR. A. Rane
Computer Engineering, Maharashtra Institute of Technology, Pune, India
2 shared publicationsRuhi Patankar
2 shared publications
Latest Publications
- 1 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Effect of Soil Attribute on Greenhouse Gas Emission from Agriculture using IoT
A farmer uses different agriculture practices to improve the crop yield generating the greenhouse gases (GHGs). The soil management practices are very important in the greenhouse gas emission from agriculture sector. In this paper we tried to find the greenhouse gas CH 4 and CO 2 emission from agriculture using soil attributes i.e. soil moisture, soil temperature, humidity, pH values, and N-P-K values. We used different machine learning algorithm such as Decision table, SOM regression, ZeroR, Randomforest, Multilayer Perception and Multiple Linear regression for comparison purpose. We collected soil sample using different sensors from the farm. We have used Mean absolute error(MAE) and Root mean squared error (RMSE) as a coefficient of correlation with 10 fold cross validation. The result shows that for CH 4 eq and CO 2 eq the SOMregressor performs better than other classifiers with perfect positive correlation whereas the ZeroR classifier gives negative correlation for CO 2 eq and CH 4 eq emission. Also the OLS statistics of multiple linear regression result shows that R squared value 0.786 means 78% predicted values are similar to actual reading of CH 4 eq emission.
- 1 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Effect of Soil Attribute on Greenhouse Gas Emission from Agriculture using IoT
A farmer uses different agriculture practices to improve the crop yield generating the greenhouse gases (GHGs). The soil management practices are very important in the greenhouse gas emission from agriculture sector. In this paper we tried to find the greenhouse gas CH 4 and CO 2 emission from agriculture using soil attributes i.e. soil moisture, soil temperature, humidity, pH values, and N-P-K values. We used different machine learning algorithm such as Decision table, SOM regression, ZeroR, Randomforest, Multilayer Perception and Multiple Linear regression for comparison purpose. We collected soil sample using different sensors from the farm. We have used Mean absolute error(MAE) and Root mean squared error (RMSE) as a coefficient of correlation with 10 fold cross validation. The result shows that for CH 4 eq and CO 2 eq the SOMregressor performs better than other classifiers with perfect positive correlation whereas the ZeroR classifier gives negative correlation for CO 2 eq and CH 4 eq emission. Also the OLS statistics of multiple linear regression result shows that R squared value 0.786 means 78% predicted values are similar to actual reading of CH 4 eq emission.
- 1 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Grapes Ripeness Estimation using Convolutional Neural network and Support Vector Machine
India is worldwide well known for exporting fruits, having a massive significance in the world. Global food security is essential for the durable production of fruits as well as for a remarkable reduction in pre and post-harvest waste. Harvesting and estimating the ripeness of fruits by a human is an expensive, laborious and time-consuming task. Ripeness estimation is carried out on single fruit like orange, apple, tomato, banana, papaya and etc., using color features. By taking into account increasing productivity of grapes and there is need to focus on ripeness estimation of grapes at the correct time. In this paper, we proposed a methodology that classifies grapes image into ripen and unripen category. A local breed of grape ‘Sonaka’ was examined during the harvest season from January to March 2019. The images were separated into two ripen categories, e.g. unripen and ripen according to the color and shape of grapes. This image was subjected to a classification model like Convolutional Neural Network (CNN) and support vector machine (SVM). Color features such as RGB and HSV and morphological features such as the shape of grapes were chosen as features for this classification model. The validation result shows that the CNN model achieves higher classification accuracy with 79.49% than the SVM classifier having 69%.
- 1 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Grapes Ripeness Estimation using Convolutional Neural network and Support Vector Machine
India is worldwide well known for exporting fruits, having a massive significance in the world. Global food security is essential for the durable production of fruits as well as for a remarkable reduction in pre and post-harvest waste. Harvesting and estimating the ripeness of fruits by a human is an expensive, laborious and time-consuming task. Ripeness estimation is carried out on single fruit like orange, apple, tomato, banana, papaya and etc., using color features. By taking into account increasing productivity of grapes and there is need to focus on ripeness estimation of grapes at the correct time. In this paper, we proposed a methodology that classifies grapes image into ripen and unripen category. A local breed of grape ‘Sonaka’ was examined during the harvest season from January to March 2019. The images were separated into two ripen categories, e.g. unripen and ripen according to the color and shape of grapes. This image was subjected to a classification model like Convolutional Neural Network (CNN) and support vector machine (SVM). Color features such as RGB and HSV and morphological features such as the shape of grapes were chosen as features for this classification model. The validation result shows that the CNN model achieves higher classification accuracy with 79.49% than the SVM classifier having 69%.
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Humanoid Robot as a Companion for the Senior Citizens
In this paper, we describe our experiment related to using a humanoid robot as a companion for the elderly people. We have analyzed the needs of elderly people and mapped our use cases in accordance with those needs. We have implemented the functionality using IBM Watson and other libraries and performed an experiment at an old age home. Our humanoid robot can detect the user's mood and engage in a friendly conversation. The experiment also indicates that social acceptance ratio of humanoid robot among elderly people is high.
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Humanoid Robot as a Companion for the Senior Citizens
In this paper, we describe our experiment related to using a humanoid robot as a companion for the elderly people. We have analyzed the needs of elderly people and mapped our use cases in accordance with those needs. We have implemented the functionality using IBM Watson and other libraries and performed an experiment at an old age home. Our humanoid robot can detect the user's mood and engage in a friendly conversation. The experiment also indicates that social acceptance ratio of humanoid robot among elderly people is high.
- 1 August 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Leaf Disease Detection and Recommendation of Pesticides Using Convolution Neural Network
Crop production problems are common in India which severely effect rural farmers, agriculture sector and the country's economy as a whole. In Crops leaf plays an important role as it gives information about the quantity and quality of agriculture yield in advance depending upon the condition of leaf. In this paper we proposed the system which works on preprocessing, feature extraction of leaf images from plant village dataset followed by convolution neural network for classification of disease and recommending Pesticides using Tensor flow technology. The main two processes that we use in our system is android application with Java Web Services and Deep Learning. We have use Convolution Neural Network with different layers five, four & three to train our model and android application as a user interface with JWS for interaction between these systems. Our results show that the highest accuracy achieved for 5-layer model with 95.05% for 15 epochs and highest validation accuracy achieved is for Slayer model with 89.67% for 20 epochs using tensor flow.
- 1 August 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Leaf Disease Detection and Recommendation of Pesticides Using Convolution Neural Network
Crop production problems are common in India which severely effect rural farmers, agriculture sector and the country's economy as a whole. In Crops leaf plays an important role as it gives information about the quantity and quality of agriculture yield in advance depending upon the condition of leaf. In this paper we proposed the system which works on preprocessing, feature extraction of leaf images from plant village dataset followed by convolution neural network for classification of disease and recommending Pesticides using Tensor flow technology. The main two processes that we use in our system is android application with Java Web Services and Deep Learning. We have use Convolution Neural Network with different layers five, four & three to train our model and android application as a user interface with JWS for interaction between these systems. Our results show that the highest accuracy achieved for 5-layer model with 95.05% for 15 epochs and highest validation accuracy achieved is for Slayer model with 89.67% for 20 epochs using tensor flow.
- Open Access
- 15 February 2017
- Published by Foundation of Computer Science in International Journal of Computer Applications
- Vol. 159 (3) , 39-42
Discrimination Aware Data Mining in Internet of Things (IoT)
- 15 February 2017
- Published by Foundation of Computer Science in International Journal of Computer Applications
- Vol. 159 (3) , 39-42
Open AccessDiscrimination Aware Data Mining in Internet of Things (IoT)
- 1 February 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 51-55
Data oriented view of a smart city: A big data approach
Smart City concept is being adopted by many governments to implement sustainable solutions and improve the quality of life of the citizens. This paper presents a data oriented view of a smart city by studying the characteristics of data generated by various sources in the city. This data would be used to build smart city components related to transport, healthcare, energy and environment. Datafication of smart city reveals that the data to be managed is Big Data. Hence, we further review how big data can be used for smart city solutions, and provide a process flow and mapping for some of the smart city components.
- 1 February 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 51-55
Data oriented view of a smart city: A big data approach
Smart City concept is being adopted by many governments to implement sustainable solutions and improve the quality of life of the citizens. This paper presents a data oriented view of a smart city by studying the characteristics of data generated by various sources in the city. This data would be used to build smart city components related to transport, healthcare, energy and environment. Datafication of smart city reveals that the data to be managed is Big Data. Hence, we further review how big data can be used for smart city solutions, and provide a process flow and mapping for some of the smart city components.
- 1 January 2017
- Published by Inderscience Publishers in International Journal of Data Science
- Vol. 2 (1) , 70
Discrimination-aware data mining: a survey
Data mining is a very important and useful technique to extract knowledge from raw data. However, there is a challenge faced by data mining researchers, in the form of potential discrimination. Discrimination means giving unfair treatment to a person just because one belongs to a minority group, without considering one's individual merit or qualification. The results extracted using data mining techniques may lead to discrimination, if a biased historical/training dataset is used. It is very important to prevent data mining technique from becoming a source of discrimination. A detailed survey of discrimination discovery methods and discrimination prevention methods is presented in this paper. This paper also presents the list of datasets used for experiments in different discrimination-aware data mining (DADM) approaches. Some ideas for future research work that may help in preventing discrimination are also discussed.
- 1 January 2017
- Published by Inderscience Publishers in International Journal of Data Science
- Vol. 2 (1) , 70
Discrimination-aware data mining: a survey
Data mining is a very important and useful technique to extract knowledge from raw data. However, there is a challenge faced by data mining researchers, in the form of potential discrimination. Discrimination means giving unfair treatment to a person just because one belongs to a minority group, without considering one's individual merit or qualification. The results extracted using data mining techniques may lead to discrimination, if a biased historical/training dataset is used. It is very important to prevent data mining technique from becoming a source of discrimination. A detailed survey of discrimination discovery methods and discrimination prevention methods is presented in this paper. This paper also presents the list of datasets used for experiments in different discrimination-aware data mining (DADM) approaches. Some ideas for future research work that may help in preventing discrimination are also discussed.
- 1 December 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 501-506
Risk assessment of X.509 certificate by evaluating Certification Practice Statements
Now a days, a lot of people and groups are using X.509 certificates to represent their identity, during online trade, so the level of purity and trustworthiness of these certificates becomes dubious. Hence, we have suggested a model which calculates the risk associated with X.509 certificates by evaluating Certificate Practice statement (CPS) document and by using certain trust criteria. For evaluating CPS document we have proposed a novel algorithm which locates certain attributes, in the CPS document. We are referring to these attributes from the prior paper of Omar and Lindsay. Our model categorizes risk in three levels-High, Medium and Low risk. It has application in detecting phishing websites which contain Https URL.
- 1 December 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 501-506
Risk assessment of X.509 certificate by evaluating Certification Practice Statements
Now a days, a lot of people and groups are using X.509 certificates to represent their identity, during online trade, so the level of purity and trustworthiness of these certificates becomes dubious. Hence, we have suggested a model which calculates the risk associated with X.509 certificates by evaluating Certificate Practice statement (CPS) document and by using certain trust criteria. For evaluating CPS document we have proposed a novel algorithm which locates certain attributes, in the CPS document. We are referring to these attributes from the prior paper of Omar and Lindsay. Our model categorizes risk in three levels-High, Medium and Low risk. It has application in detecting phishing websites which contain Https URL.
- 24 November 2016
- Published by Springer Nature
- p. 372-389
Risk Evaluation of X.509 Certificates – A Machine Learning Application
- 24 November 2016
- Published by Springer Nature
- p. 372-389
Risk Evaluation of X.509 Certificates – A Machine Learning Application