“Ashish has been eager to learn new skills, including ones in my field of data science. I'm familiar with his work at Packt within my industry, as well; he is very thorough in his reviews. If you need someone who is eager to learn and collaborate on projects, I would recommend reaching out to Ashish about your project! If he doesn't know the area yet, he'll figure out what he needs to know quickly and reach out to those in the field to learn the new material.”
About
I have over 12+ years, Author, Data Scientist and Researcher with 8+ Years of Experience…
Experience
Education
-
2021 Qiskit Global Summer School on Quantum Machine Learning
-
Activities and Societies: Completed the two-week intensive course provided by IBM Quantum, completing all graded lab work assignments with a final cumulative score above 75%, demonstrating applied understanding and comfort with and about Quantum Computing and Quantum Machine Learning using Qiskit
• Lab 1: Quantum Computing Operation and Algorithms
• Lab 2: Variational Algorithms
• Lab 3: Quantum Feature Maps, kernels and Support Vector Machines:
• Lab 4: Training Quantum Circuits:
• Lab 5: Training Quantum Circuits
• Links for more topics : https://github.com/ashishpatel26/IBM-Quantum-Machine-Learning-2021 -
-
Activities and Societies: Semester 1 Classical Computing, Quantum Computing in the Abstract, Math: Introduction to Vectors and Complex Numbers, Probability, Math for Quantum Mechanics, Introduction to Python Programming Semester 2 Quantum Mechanics, The Qubit and Bloch Sphere, Gates, Measurements and Quantum Circuits, Quantum Key Distribution, Superdense Coding + Quantum Teleportation, Classic Algorithms, Deutsch-Josza Algorithm, Grover’s Algorithm, VQE & QAOA, Metrics, and Implementation
Train the future diverse quantum workforce
● Introduce students to the field of quantum computing
● Develop foundational skills, including math, computer science, and physics, necessary to pursue quantum computing
● Prepare students with tangible and real-world STEM skills
● Deepen understanding of quantum applications
● Learn about career opportunities in quantum
● Increase diversity in STEM fields
● Introduce students to industry and academic leaders in quantum…Train the future diverse quantum workforce
● Introduce students to the field of quantum computing
● Develop foundational skills, including math, computer science, and physics, necessary to pursue quantum computing
● Prepare students with tangible and real-world STEM skills
● Deepen understanding of quantum applications
● Learn about career opportunities in quantum
● Increase diversity in STEM fields
● Introduce students to industry and academic leaders in quantum
● Form a global cohort of future quantum leaders
-
-
Activities and Societies: Media Team in College Activities,etc..
Dissertation titled “Quantify the severity of Jellyfish Attack on Mobile adhoc Network(MANET)”, 2015. Published in International Journal of Innovative Research in Technology, May 2015.
-
-
B.E in Information Technology
Licenses & Certifications
-
-
-
-
What is the Metaverse?
Meta
IssuedCredential ID https://coursera.org/share/93f11d6350b24fc7367f705d7bf8e252 -
-
IBM Quantum Summer School 2022
IBM
IssuedCredential ID https://www.credly.com/badges/3689701c-bf7f-43cc-9caa-30ff2be2fbe4 -
IBM Quantum Summer Spring Challenge 2022
IBM
IssuedCredential ID https://challenges.quantum-computing.ibm.com/spring-2022 -
-
-
-
Custom Models, Layers, and Loss Functions with TensorFlow
Coursera | Google
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Volunteer Experience
-
Volunteer
YUVA Unstoppable
- Present 9 years 7 months
Education
Teach a poor people's children,who don't have enough money or food to live.Gives stationaries to child so they can learn with us. Finally with yuva unstoppable,We are trying to Give good education for this children so no single child can left without education.
-
Professional Speaker
AWS Ahmedabad Community
- Present 1 year 9 months
Education
In December 2022, an AWS MLOps session unveiled revolutionary advancements in Machine Learning Operations. It addressed challenges in ML model operationalization, offering a scalable infrastructure. AWS SageMaker simplified the ML workflow, from data prep to model deployment. AWS Step Functions provided seamless orchestration of ML pipelines. Real-life use cases showcased the business value of MLOps, optimizing customer engagement and supply chain operations. AWS CodeGuru ensured model…
In December 2022, an AWS MLOps session unveiled revolutionary advancements in Machine Learning Operations. It addressed challenges in ML model operationalization, offering a scalable infrastructure. AWS SageMaker simplified the ML workflow, from data prep to model deployment. AWS Step Functions provided seamless orchestration of ML pipelines. Real-life use cases showcased the business value of MLOps, optimizing customer engagement and supply chain operations. AWS CodeGuru ensured model governance, while Sagemaker Ground Truth ensured data accuracy. AWS Elastic Inference optimized ML inference, and hands-on workshops empowered participants. The session marked a milestone in ML Ops, empowering organizations to harness the transformative potential of AI.
-
Guest Speaker
Google Developers Group Gandhinagar
- 1 month
Education
Recently, I immersed myself in a transformative MLOps session at Google Developers Group Gandhinagar. Explored cutting-edge practices in version control, continuous integration, and automated deployment pipelines. 🚀 Excited to leverage these insights for seamless machine learning lifecycle management! 💡 #MLOps #GoogleDevelopersGroup…
Recently, I immersed myself in a transformative MLOps session at Google Developers Group Gandhinagar. Explored cutting-edge practices in version control, continuous integration, and automated deployment pipelines. 🚀 Excited to leverage these insights for seamless machine learning lifecycle management! 💡 #MLOps #GoogleDevelopersGroup #MachineLearning
https://www.linkedin.com/posts/gdggandhinagar_mlops-datascience-techspeaker-activity-7138922325860683776-eY78?utm_source=share&utm_medium=member_ios
Publications
-
Semantic segmentation approaches for crop classification with multi-altitude Google Earth imagery
Journal of Integrated Science and Technology
Abstract
Image Segmentation with Remote Sensing (RS) dataset is highly challenging in many applications, such as agriculture, land classification, and disaster management, due to high interclass similarities in features, noise in the data, and the need for appropriate algorithms for segmentation. Traditional approaches utilize machine learning (ML) and deep learning (DL) algorithms, and they have performed well for this in the availability of many image datasets. Because of the…Abstract
Image Segmentation with Remote Sensing (RS) dataset is highly challenging in many applications, such as agriculture, land classification, and disaster management, due to high interclass similarities in features, noise in the data, and the need for appropriate algorithms for segmentation. Traditional approaches utilize machine learning (ML) and deep learning (DL) algorithms, and they have performed well for this in the availability of many image datasets. Because of the advantages and applications of U-Net architecture for semantic segmentation, even with the scarcity of training image samples, this paper aims to apply pixel-wise classification of aerial images and image segmentation with U-Net and U-Net image blocks. The paper's novelty is to apply segmentation for a single-day varying altitude image dataset collected from a village in Gujarat using the Google Earth’s sentinel satellite image views. The classification has been carried out for two major crops, wheat, and Ricinus. The assessment of various architectural frameworks for the segmentation and classification, including ML, DL, and U-Net, by fine-tuning the models with monotonic learning rate (LR) and Cyclic LR, are included in this paper. The U-Net architecture for a dataset with 500m altitudes with a Validation Accuracy of 94.6%, Loss of 3.48%, Area Under Cover (AUC) of 95.76%, Sensitivity of 99.5%, and Specificity of 99.07% and another U-Net image block ResUNet architecture with Cyclic LR for 1000 m altitude outperform the traditional ML, DL algorithms with Validation Accuracy 98.5%, Loss 0.93%, AUC 88.91%, Sensitivity 92.86%, and Specificity 95.42%. Significantly, the U-Net and its image block architectures with Cyclic LR outperform the ML and DL variants.Other authorsSee publication -
Monocular 3D Object Detection using Multi-Stage Approaches with Attention and Slicing aided hyper inference
Arxiv
3D object detection is vital as it would enable us to capture objects' sizes, orientation, and position in the world. As a result, we would be able to use this 3D detection in real-world applications such as Augmented Reality (AR), self-driving cars, and robotics which perceive the world the same way we do as humans. Monocular 3D Object Detection is the task to draw 3D bounding box around objects in a single 2D RGB image. It is localization task but without any extra information like depth or…
3D object detection is vital as it would enable us to capture objects' sizes, orientation, and position in the world. As a result, we would be able to use this 3D detection in real-world applications such as Augmented Reality (AR), self-driving cars, and robotics which perceive the world the same way we do as humans. Monocular 3D Object Detection is the task to draw 3D bounding box around objects in a single 2D RGB image. It is localization task but without any extra information like depth or other sensors or multiple images. Monocular 3D object detection is an important yet challenging task. Beyond the significant progress in image-based 2D object detection, 3D understanding of real-world objects is an open challenge that has not been explored extensively thus far. In addition to the most closely related studies.
Other authorsSee publication -
ConvMax Classification of COVID-19, pneumonia, and normal lungs from X-ray images using CNN with modified max-pooling layer
Taylor and Francis Group
The coronavirus disease (COVID) erupted at the end of 2019 and quickly spread throughout the globe. The virus is capable of spreading at an exponential speed within a few days. As COVID-19 is transmissible among humans, it is required to detect infectious humans as soon as possible. One of the rapid techniques to identify COVID-19 infection in human beings is an X-ray of various body parts. As COVID-19 cases rise, computerized diagnostic methods are needed for initial recognition and prognosis…
The coronavirus disease (COVID) erupted at the end of 2019 and quickly spread throughout the globe. The virus is capable of spreading at an exponential speed within a few days. As COVID-19 is transmissible among humans, it is required to detect infectious humans as soon as possible. One of the rapid techniques to identify COVID-19 infection in human beings is an X-ray of various body parts. As COVID-19 cases rise, computerized diagnostic methods are needed for initial recognition and prognosis of infection. In this study, the convolution neural network deep net architecture is developed for classifying among binary classes – COVID-19 and normal – and for multi-classes – COVID-19, pneumonia, and normal chest X-ray images. The novelty is added in terms of larger features for map extraction. When a 3 × 3 size max-pooling layer is applied, it extracts more knowledge than the 2 × 2 size max-pooling layer, which helps increase the model knowledge and performance. The proposed model is estimated to deliver accurate indicatory results of classification accuracy.
Other authorsSee publication -
Deep Learning Model for Acoustics Signal Based Preventive Healthcare Monitoring and Activity of Daily Living
IEEE
To cope with the increasing healthcare costs and nursing shortages in the Aging Society the care system is transferred, as much as possible, to the home environment, making use of ambient assisted living (AAL) monitoring and communication possibilities and to actively involve informal cares to fill in large part of the care that is needed. The proposed system is the AAL based, acoustics sensing system ready to dissect, recognize, and distinguish specific acoustic events occurring in day-by-day…
To cope with the increasing healthcare costs and nursing shortages in the Aging Society the care system is transferred, as much as possible, to the home environment, making use of ambient assisted living (AAL) monitoring and communication possibilities and to actively involve informal cares to fill in large part of the care that is needed. The proposed system is the AAL based, acoustics sensing system ready to dissect, recognize, and distinguish specific acoustic events occurring in day-by-day life situations, which empowers not only the individual subjects but also the healthcare professionals to remotely follow the status of each individual continuously. This system only processes the background acoustics related to the activity of daily living (ADL) for preventive healthcare. The novel contribution of the research is based on prototype development, audio signal processing algorithms and deep learning algorithms to satisfy the research gap.
Other authorsSee publication -
Hands-on Time Series Analysis with Python
Apress Springer Nature
Introduction
This book explains the concepts of time series from traditional to bleeding-edge techniques with full-fledged examples.
The book begins by covering time series fundamentals and its characteristics, the structure of time series data, pre-processing, and ways of crafting the features through data wrangling. Next, it covers the traditional time series techniques like Smoothing methods, ARMA, ARIMA, SARIMA, SARIMAX, VAR, VARMA using trending framework like StatsModels…Introduction
This book explains the concepts of time series from traditional to bleeding-edge techniques with full-fledged examples.
The book begins by covering time series fundamentals and its characteristics, the structure of time series data, pre-processing, and ways of crafting the features through data wrangling. Next, it covers the traditional time series techniques like Smoothing methods, ARMA, ARIMA, SARIMA, SARIMAX, VAR, VARMA using trending framework like StatsModels, pmdarima. Further, Book explains the building classification models using sktime, and covers how to leverage advance deep learning-based techniques like ANN, CNN, RNN, LSTM, GRU and Autoencoder to solve time series problem using Tensorflow. It finally concludes by explaining the popular framework fbprophet for modeling time series analysis.
After completion of the book, the reader will have a good understanding of working with different techniques of time series methods. All the codes presented in this notebook are available in Jupyter notebooks, which allows readers to do hands-on and enhance them in exciting ways.
What You'll Learn
Explains basics to advanced concepts of time series
How to design, develop, train, and validate time-series methodologies
What are smoothing, ARMA, ARIMA, SARIMA,SRIMAX, VAR, VARMA techniques in time series and how to optimally tune parameters to yield best results
Learn how to leverage bleeding-edge techniques such as ANN, CNN, RNN, LSTM, GRU, Autoencoder to solve both Univariate and multivariate problems by using two types of data preparation methods for time series.
Univariate and multivariate problem solving using fbprophet.
Other authorsSee publication -
Proposal and Preliminary Fall-related Activities Recognition in Indoor Environment
IEEE
Falls are a noteworthy reason for grievances and deaths in elderlies. Notwithstanding when no damage happens, about majority of elderlies are identity unfit to get up without help. The expanded time of lying on the floor frequently prompts restorative complications, including muscle impairment, lack of hydration, unease, and trepidation of falling. Here, a fall sensing unit is accounted that is affixed to a subjects' midsection and incorporates a 3-axis accelerometer, 3-axis gyroscope, a…
Falls are a noteworthy reason for grievances and deaths in elderlies. Notwithstanding when no damage happens, about majority of elderlies are identity unfit to get up without help. The expanded time of lying on the floor frequently prompts restorative complications, including muscle impairment, lack of hydration, unease, and trepidation of falling. Here, a fall sensing unit is accounted that is affixed to a subjects' midsection and incorporates a 3-axis accelerometer, 3-axis gyroscope, a multiplexer, a filter, and a microcontroller. Moreover, the fall detection system also used IMU data on the mobile phone. Change in angular velocity, noise cancelation, and the ADC transformation was achieved by the hardware. The handled flag is conveyed to a PC through ZigBee and processed through the dedicated programming. Fall sensing approach comprised feature selection, mining and a machine learning calculation for characterizing the parameters. In this paper, we propose a fall discovery calculation which is shaped by feature selection, discovery, mining and handling. An aggregate of six highlights was ascertained in feature selection. Four of them are identified with the gravity vector which is extricated from accelerometer information by utilizing the low-pass filter. As falling generally happens in a vertical course, the gravity-related characteristics are helpful. The system also uses one of the ambient sensing units, which is a movement sensing unit. The PIR sensor-based movement sensing unit is used to enhance the accuracy of fall detection activity. The feature from the movement sensing unit substantially reduced the false alarms.
Other authorsSee publication -
A SURVEY ON FACIAL EXPRESSION RECOGNITION USING DEEP LEARNING APPROACH
INTERNATIONAL JOURNAL OF NOVEL RESEARCH AND DEVELOPMENT
n the course of the most recent couple of years, profound neural systems have gotten the most consideration insoftware engineering, particularly in design acknowledgment, machine vision and machine learning. One of its superb applications is in the feeling acknowledgment by means of Facial expression recognition region. Facial expression recognition investigation is valuable for some undertakings and the utilization of profound learning here is…
n the course of the most recent couple of years, profound neural systems have gotten the most consideration insoftware engineering, particularly in design acknowledgment, machine vision and machine learning. One of its superb applications is in the feeling acknowledgment by means of Facial expression recognition region. Facial expression recognition investigation is valuable for some undertakings and the utilization of profound learning here is additionally growing quick. We survey some current research works in this space, present some new applications and demonstrate the general strides to actualizing every one of them
-
A SURVEY ON DEEP LEARNING METHOD USED FOR CHARACTER RECOGNITION
International Journal Of Creative and Innovative Research In All Studie
The field of Artificial Intelligence is very fashionable today, especially neural networks that work well in various areas such as speech recognition and natural language processing. This Research Article briefly describes how deep learning models work and what different techniques are used in text recognition. It also describes the great progress that has been made in the field of medicine, the analysis of forensic documents, the recognition of license plates, banking, health and the legal…
The field of Artificial Intelligence is very fashionable today, especially neural networks that work well in various areas such as speech recognition and natural language processing. This Research Article briefly describes how deep learning models work and what different techniques are used in text recognition. It also describes the great progress that has been made in the field of medicine, the analysis of forensic documents, the recognition of license plates, banking, health and the legal industry. The recognition of handwritten characters is one of the research areas in the field of artificial intelligence. The individual character recognition has a higher recognition accuracy than the complete word recognition. The new method for categorizing Freeman strings is presented using four connectivity events and eight connectivity events with a deep learning approach.
-
EFFECTIVE CREDIT DEFAULT SCORING USING ANOMALY DETECTION
International Journal of Science Technology & Engineering
In recent years there has been a trend towards online purchase so stealing of credit data is high like identity of the credit card owner, password or etc. the attacker may use this data for to take loan from financial domain and they make credit default. Credit scoring is the give the creditworthiness of person. Anomaly Detection is the process of classifies unusual behavior. It is important data analysis task used for classify interesting and emerging patterns, trends and anomalies from data…
In recent years there has been a trend towards online purchase so stealing of credit data is high like identity of the credit card owner, password or etc. the attacker may use this data for to take loan from financial domain and they make credit default. Credit scoring is the give the creditworthiness of person. Anomaly Detection is the process of classifies unusual behavior. It is important data analysis task used for classify interesting and emerging patterns, trends and anomalies from data. Anomaly detection is an important tool to detect irregularity in many different domains including financial fraud detection, computer network intrusion, human behavioral analysis and many more. In today’s era the credit and Loan Default is become high because of many fraudulent activity or increase online purchases. To perform anomaly detection in this paper linear regression with rule based classification and logistic regression is used. The preprocessing is used for to perform explore, analyze and determine the factor that play crucial role to find credit default.
-
TAG BASED IMAGE RETRIEVAL USING NATURAL LANGUAGE PROCESSING(NLP)
International Journal of Advance Research and Innovative Ideas in Education
The Field of Natural Language Processing (NLP) is getting to be one of the dynamic regions in Human-PC connection which has seen a study moved in both research and philosophy bearing in the previous couple of years. Normal Language is advantageous technique for getting to information, particularly for easy going clients who don't comprehend the specialized method for seeking. To translate the client inquiry which is communicated in type of inquiries, sentence and irregular words utilizing…
The Field of Natural Language Processing (NLP) is getting to be one of the dynamic regions in Human-PC connection which has seen a study moved in both research and philosophy bearing in the previous couple of years. Normal Language is advantageous technique for getting to information, particularly for easy going clients who don't comprehend the specialized method for seeking. To translate the client inquiry which is communicated in type of inquiries, sentence and irregular words utilizing characteristic dialect handling are present pattern. This idea of common dialect inquiry handling assists the web crawlers with retrieving more semantic pictures identified with the client question. Flickr is one of the well-known informal communication tag based picture seek benefit which recover the pictures in type of labels, which is not generally splendidly identifying with information gave by individuals who hunt to them.
-
A SURVEY ON TAG BASED IMAGE RETRIEVAL USING NATURAL LANGUAGE PROCESSING
International Journal of Innovative Research in Technology
The Field of Natural Language Processing (NLP) is getting to be one of the dynamic regions in Human-PC connection which has seen a study moved in both research and philosophy bearing in the previous couple of years. Normal Language is advantageous technique for getting to information, particularly for easy going clients who don't comprehend the specialized method for seeking. To translate the client inquiry which is communicated in type of inquiries, sentence and irregular words utilizing…
The Field of Natural Language Processing (NLP) is getting to be one of the dynamic regions in Human-PC connection which has seen a study moved in both research and philosophy bearing in the previous couple of years. Normal Language is advantageous technique for getting to information, particularly for easy going clients who don't comprehend the specialized method for seeking. To translate the client inquiry which is communicated in type of inquiries, sentence and irregular words utilizing characteristic dialect handling are present pattern. This idea of common dialect inquiry handling assists the web crawlers with retrieving more semantic pictures identified with the client question. Flickr is one of the well-known informal communication tag based picture seek benefit which recover the pictures in type of labels, which is not generally splendidly identifying with information gave by individuals who hunt to them.
-
A SURVEY OF ANOMALY DETECTION TECHNIQUE FOR CREDIT DEFAULT
International Journal of Innovative Research in Technology
Anomaly Detection is the process of classify unusual behavior. It is important data analysis task used for classify interesting and emerging patterns, trends and anomalies from data. Anomaly detection is an important tool to detect irregularity in many different domains including financial fraud detection, computer network intrusion, human behavioral analysis and many more. In today’s era the credit and Loan Default is become high because of many fraudulent activity or increase online…
Anomaly Detection is the process of classify unusual behavior. It is important data analysis task used for classify interesting and emerging patterns, trends and anomalies from data. Anomaly detection is an important tool to detect irregularity in many different domains including financial fraud detection, computer network intrusion, human behavioral analysis and many more. In today’s era the credit and Loan Default is become high because of many fraudulent activity or increase online purchases. This paper give survey about many technique for anomaly detection in credit or loan default.
-
DETECTION AND PREVENTION TECHNIQUE FROM JELLYFISH DELAY VARIANCE ATTACK
International Journal of Innovative Research in Technology
An ad hoc network is anaccumulation of wireless mobile nodes dynamically forming a tentative network without the use of any existing network infrastructure or medication administration. Due to the limited transmission range of wireless network interfaces, multiple network "hops" may be needed for one node to transpositionof data with another node across the network. In recent years, wireless ad hoc networks (WANETs) have become very in-vogue due to their wide range of petition and their ability…
An ad hoc network is anaccumulation of wireless mobile nodes dynamically forming a tentative network without the use of any existing network infrastructure or medication administration. Due to the limited transmission range of wireless network interfaces, multiple network "hops" may be needed for one node to transpositionof data with another node across the network. In recent years, wireless ad hoc networks (WANETs) have become very in-vogue due to their wide range of petition and their ability to be deployed under normal and rugged conditions while supporting high data rates. Although many intrusion detection and trust-based systems have been developed to protect ad hoc networks against misconduct such as rushing attacks, query-flood attacks, and selfishness of nodes, these aegis mechanisms are still not able to detect protocol compliant attacks called Jellyfish (JF) attacks. They target closed-loop flows such as TCP that are responsive to network conditions like delay and packet losses and can easily partition the network. In this paper, we introduce a technique which can be used to detect and detain Jellyfish delay variance attacks in ad hoc networks
-
ENERGY EFFICIENT DSR TO IMPROVE THE ROUTING IN MANET
International Journal of Innovative Research in Technology
Dynamic Source Routing protocol (DSR) has been accepted itself as one of the distinguished and dominant routing protocols for Mobile Ad Hoc Networks (MANETs). From various performance analysis and results, it is shown that DSR has been an outstanding routing protocol that outperforms consistently than any other routing protocols. But it could not pervade the same place when the performance was considered in term of energy consumption at each node, energy consumption of the networks, energy…
Dynamic Source Routing protocol (DSR) has been accepted itself as one of the distinguished and dominant routing protocols for Mobile Ad Hoc Networks (MANETs). From various performance analysis and results, it is shown that DSR has been an outstanding routing protocol that outperforms consistently than any other routing protocols. But it could not pervade the same place when the performance was considered in term of energy consumption at each node, energy consumption of the networks, energy consumption per successful packet transmission, and energy consumption of node due to different overhead. Because, DSR protocol does not take energy as a parameter into account at all. And as MANET is highly sensible towards the power related issues and energy consumption as it is operated by the battery with the limited sources, needed to be used efficiently, so that the life time of the network can be prolonged and performance can be enhanced. we have proposed a novel energy efficient DSR (Dynamic Source Routing) routing protocol which modified to improve the networks lifetime in MANET in terms of energy. We also provided a solution, based on considering the energy of each node because each node's energy state has a huge influence on the entire network lifetime. All simulation is performed in NS2 simulator.
-
A SURVEY OF ASSOCIATION RULE BASED TECHNIQUES FOR PRESERVING PRIVACY AND SECURITY
International Journal of Innovative Research in Technology
Data mining is that the extraction of attention-grabbing patterns or data from large quantity of information. In recent years, with the explosive development in web, knowledge storage and processing technologies, privacy preservation has been one in every of the larger issues in data processing. variety of ways and techniques are developed for privacy conserving data processing. This paper provides a good survey of various privacy conserving data processing algorithms and analyses the…
Data mining is that the extraction of attention-grabbing patterns or data from large quantity of information. In recent years, with the explosive development in web, knowledge storage and processing technologies, privacy preservation has been one in every of the larger issues in data processing. variety of ways and techniques are developed for privacy conserving data processing. This paper provides a good survey of various privacy conserving data processing algorithms and analyses the representative techniques for privacy conserving data processing, and points out their deserves and demerits. Finally the current issues and directions for future analysis area unit mentioned.
Courses
-
M.E
Computer Engineering
Projects
-
Audio/ Sound Classification with Environment Sounds Dataset
-
The ESC-50 dataset is a labeled collection of 2000 environmental audio recordings suitable for bench marking methods of environmental sound classification. The dataset consists of 5-second-long recordings organized into 50 semantical classes (with 40 examples per class) loosely arranged into 5 major categories. I trained Convolution Neural Network for sound classification. I achieved classification accuracy of approx ~83%. MFCC (mel-frequency cepstrum) feature is used to train models. Other…
The ESC-50 dataset is a labeled collection of 2000 environmental audio recordings suitable for bench marking methods of environmental sound classification. The dataset consists of 5-second-long recordings organized into 50 semantical classes (with 40 examples per class) loosely arranged into 5 major categories. I trained Convolution Neural Network for sound classification. I achieved classification accuracy of approx ~83%. MFCC (mel-frequency cepstrum) feature is used to train models. Other features like short term fourier transform, chroma, melspectrogram can also be extracted. This dataset is challenged with 50 classes related to accuracy of Audio prediction. Achieved 83% accuracy with mapping weight technique of previous maturity of model. I got prediction 80% correct.
-
Image Classification
-
The largest memory manufacturer wants to classify defective and non-defective images with high accuracy and a lower overkill rate.
In this project, the client provided 10K images containing defective and non-defective images. The main objective of the project is to classify the two type of images and store the result. Project required to read massive images of the network path defined in the configuration. We have developed system which runs every hour and reads the files from the…The largest memory manufacturer wants to classify defective and non-defective images with high accuracy and a lower overkill rate.
In this project, the client provided 10K images containing defective and non-defective images. The main objective of the project is to classify the two type of images and store the result. Project required to read massive images of the network path defined in the configuration. We have developed system which runs every hour and reads the files from the network. These files are passed into the trained model to predict the result. The results are saved in two formats, one in a flat file and the other in the database, which can be used to generate reports.
The challenge was to identify defective images with a high accuracy of nearly 99%. As a projectile, we need more images to form the deep learning model generated by the image enhancement technique. In a second step, read the image using Open CV and apply the learning transfer to develop the model.
As a result, our image classification model obtained a satisfactory default image classification with a Type 2 error of zero.
Responsibilities:
• Collection of requirements of the Business team.
• Increase training data using the image augmentation technique.
• Used the concept of transfer learning to obtain good accuracy.
• Implementation of the Deep Learning Image Classification Model.
• Model validated in thousands of images.
• Implementation of the model in GPU.
• Weekly Scrum meeting.
-
News analytic to predict stock price performance(Finance Assets)
-
The ubiquity of information today empowers investors at any scale to make better investment decisions. Challenge in this project to ingesting and interpreting the data to figure out which information is helpful, finding the flag in this ocean of data. By investigating news data to anticipate(predict) stock prices, and the condition of research in understanding the prescient power of the news. harnessed, to help predict financial outcomes and produce critical monetary effect everywhere…
The ubiquity of information today empowers investors at any scale to make better investment decisions. Challenge in this project to ingesting and interpreting the data to figure out which information is helpful, finding the flag in this ocean of data. By investigating news data to anticipate(predict) stock prices, and the condition of research in understanding the prescient power of the news. harnessed, to help predict financial outcomes and produce critical monetary effect everywhere throughout the world. This undertaking depends on news data affect so we dissect the sentiments first using TFIDF(NLP Technique) and other NLP techniques which generate features and check their effect on stocks.
Responsibilities:
• Requirement gathering from the Business Team.
• Gather market and news data.
• Explore data to understand the sentiments.
• Generate feature of news data using TFIDF.
• Implemented and Designed different parameter for ML model.(Microsoft Designed Light GBM)
• Model validation/Deployment.
-
Predictive Maintenance
-
In this project one of leading automotive manufacture wants to detect When their robots need a maintenance before shutdown. So based on problem they can save their build an LSTM network in order to predict remaining useful life (or time to failure) of robots. The network uses simulated robot’s sensor values to predict when an robot will fail in the future so that maintenance can be planned in advance.
Time Series Analysis : How many more cycles an in-service robot will last before it…In this project one of leading automotive manufacture wants to detect When their robots need a maintenance before shutdown. So based on problem they can save their build an LSTM network in order to predict remaining useful life (or time to failure) of robots. The network uses simulated robot’s sensor values to predict when an robot will fail in the future so that maintenance can be planned in advance.
Time Series Analysis : How many more cycles an in-service robot will last before it fails?
Binary classification: Is this robot going to fail within number of cycles?
Responsibilities:
• Requirement gathering from the Business Team.
• Gathering data.
• Used sequence to sequence Modeling approach.
• Training LSTM model.
• Model validation/Deployment.
-
Property recommendation
-
Real estate firm wants property recommendation to dealers. Client has large number of property data and wanted recommendation system to their dealers. Data is sparse, so It suffer from scalability issue which is solved using cluster approach and new listed property doesn’t get much attention which is cold start problem and solved by designing hybrid similarity approach.
Responsibilities:
• Requirement gathering from the Business Team.
• Matrix Generation from raw…Real estate firm wants property recommendation to dealers. Client has large number of property data and wanted recommendation system to their dealers. Data is sparse, so It suffer from scalability issue which is solved using cluster approach and new listed property doesn’t get much attention which is cold start problem and solved by designing hybrid similarity approach.
Responsibilities:
• Requirement gathering from the Business Team.
• Matrix Generation from raw data.
• Identify and solved cold-start problem
• Solved sparse data issue using clustering approach.
• Implemented and Designed different similarity algorithm and recommendation.
• Model validation/Deployment.
Honors & Awards
-
Kaggle Master World Rank 3
Kaggle
* Kaggle Kernel Master World Rank3
* Three time potential Kernel Award
Languages
-
English
-
-
Hindi
-
-
Gujarati
-
Recommendations received
9 people have recommended Ashish
Join now to viewOther similar profiles
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore MoreOthers named Ashish Patel 🇮🇳 in India
-
Ashish Patel
Global Leadership Development Strategist | Executive Coach | Organizational Excellence Consultant | Managing Director & Founder Consultant, Euphoria Consulting
-
Ashish Patel
Data science/Data analyst|KRMU student Executives boot camps IIT mandi Electronics and information technology
-
Ashish Patel
Data Scientist | WTW | IIT Bombay (2021-23)
-
Ashish Patel
Graduate Student in San Diego State University | MS in Big Data Analytics Fall 2024 |
4669 others named Ashish Patel 🇮🇳 in India are on LinkedIn
See others named Ashish Patel 🇮🇳