Abhishek Kumar

Abhishek Kumar

Patna, Bihar, India
836 followers 500+ connections

Join to follow

About

Currently working as Data Scientist in the domain of Health Care.

Activity

Join now to see all activity

Experience

  • Highbrow Technology Inc Graphic

    Highbrow Technology Inc

  • -

    India

  • -

    Hyderabad, Telangana, India

  • -

    India

  • -

    Bihta, Bihar, India

  • -

    Kolkata, West Bengal, India

  • -

    Kolkata Area, India

  • -

    Hyderabad Area, India

  • -

    Hyderabad Area, India

  • -

    Lucknow Area, India

  • -

    PATNA

Education

Licenses & Certifications

Publications

  • A Multi-task Multi-modal Framework for Sentiment and Emotion aided Cyberbully Detection

    IEEE

    With the expansion of digital sphere and advancement of technology, cyberbullying has become increasingly common, especially among teenagers. In this work, we have created a benchmark Hindi-English code-mixed corpus called BullySentEmo, annotated with bully, sentiment and emotion labels for investigating how sentiment and emotion label information can help in identifying cyberbully in a better way. For a vast portion of India, both of these languages constitute the primary means of…

    With the expansion of digital sphere and advancement of technology, cyberbullying has become increasingly common, especially among teenagers. In this work, we have created a benchmark Hindi-English code-mixed corpus called BullySentEmo, annotated with bully, sentiment and emotion labels for investigating how sentiment and emotion label information can help in identifying cyberbully in a better way. For a vast portion of India, both of these languages constitute the primary means of communication, and language mixing is common in everyday speech. A multi-task multi-modal framework called MT-MM-Bert+VecMap based on two different embedding schemes for the efficient representations of code-mixed data with emoji modality has been developed. Our proposed multitask-multimodal framework outperforms all the single-task and unimodal baselines with the highest accuracy values of 82.05(+/- 1.36)%, 77.87(+/- 1.93)% and 58.05(+/-2.78) for the Cyberbully detection(CBD) task, Sentiment analysis(SA) task and Emotion recognition(ER) task, respectively.

    See publication

  • Covid 19 Prediction from X Ray Images Using Fully Connected Convolutional Neural Network

    Accociation for Computing Machinery

    COVID 19 pandemic has paralyzed the whole world irrespective of any discrimination. To contain the infection effective testing of people plays a vital role. Usually, chest X-ray image-based diagnosis using manual methods is carried out, which is not only time-consuming but also paves way for asymptomatic patients to transmit the virus at a faster pace. Chest X-ray image analysis using a fully connected convolutional neural network (CNN) has been proposed in this paper to solve the purpose. The…

    COVID 19 pandemic has paralyzed the whole world irrespective of any discrimination. To contain the infection effective testing of people plays a vital role. Usually, chest X-ray image-based diagnosis using manual methods is carried out, which is not only time-consuming but also paves way for asymptomatic patients to transmit the virus at a faster pace. Chest X-ray image analysis using a fully connected convolutional neural network (CNN) has been proposed in this paper to solve the purpose. The fully connected CNN with two variants of convolution especially DSC has proved its efficiency in detecting COVID 19 infections.

    See publication

Projects

  • Apparel Recommendation

    A recommendation engine which suggests similar apparel to the given apparel over amazon website (here taken for example).

    See project

  • Facebook Friend Recommendation

    This project is about predicting future friends to any user, it is based on graph mining.

    See project

  • Quora question Pair Similarity Problem

    Quora is a place to gain and share knowledge—about anything. It’s a platform to ask questions and connect with people who contribute unique insights and quality answers. Problem statement was to identify which questions asked on Quora are duplicates of questions that have already been asked. This could be useful to instantly provide answers to questions that have already been answered. We are tasked with predicting whether a pair of questions are duplicates or not.

    See project

  • Self Driving Car

    A small project to experiment different combination of parameter to a Self Driving Car.

    See project

  • ltfs-datascience-finhack-an-online-hackathon

    It was an online Hackathon. The purpose of this analysis is to make up a prediction model where we will be able to accurately predict the probability of loanee/borrower defaulting on a vehicle loan in the first EMI (Equated Monthly Instalments) on the due date.

    See project

  • Stackoverflow-Tag-Predictor

    Suggest the tags based on the content that was there in the question posted in Stackoverflow. It is a multilabel classification problem. Performance metric is Micro-Averaged F1-Score (Mean F Score) : The F1 score can be interpreted as a weighted average of the precision and recall, where an F1 score reaches its best value at 1 and worst score at 0. The relative contribution of precision and recall to the F1 score are equal.

    See project

  • Human Activity Detection

    This project is to build a model that predicts the human activities such as Walking, Walking Upstairs, Walking Downstairs, Sitting, Standing or Laying. Accelerometer and Gyroscope readings are taken from 30 volunteers (referred as subjects) while performing the above 6 Activities. Given a new datapoint we have to predict the activity, is our problem statement. Accuracy is the performance matrix as the dataset is balanced.

    See project

  • Personalized-Cancer-Diagnosis

    Our objective was the classification of Cancer Classes based upon the Gene, Variation & Text and the problem statement was to classify the given genetic variations/mutations based on evidence from text-based clinical literature. Various ML model were applied and the results were compared. We've taken Random model as base model to judge the performance of other models with log-loss as performance matrix because of the business requirements. Here, we want to know the probability of belonging to…

    Our objective was the classification of Cancer Classes based upon the Gene, Variation & Text and the problem statement was to classify the given genetic variations/mutations based on evidence from text-based clinical literature. Various ML model were applied and the results were compared. We've taken Random model as base model to judge the performance of other models with log-loss as performance matrix because of the business requirements. Here, we want to know the probability of belonging to each class for every data point. Best results were achieved in case of Logistic Regression with BoW and TF-IDF (with both balanced and unbalanced class) with unigram and bigrams.

    See project

  • Taxi demand prediction in New York City

    Our main objective is to predict number of pickups, given location co-ordinates (latitude and longitude) and time, in the query region and surrounding regions. And to solve this we would be using data collected in Jan 2015 to predict the pickups in Jan - Mar 2016. Main attraction of this project was use Fourier transform as feature. We have chosen our error metric for comparison between models as MAPE (Mean Absolute Percentage Error) so that we can know that on an average how good is our model…

    Our main objective is to predict number of pickups, given location co-ordinates (latitude and longitude) and time, in the query region and surrounding regions. And to solve this we would be using data collected in Jan 2015 to predict the pickups in Jan - Mar 2016. Main attraction of this project was use Fourier transform as feature. We have chosen our error metric for comparison between models as MAPE (Mean Absolute Percentage Error) so that we can know that on an average how good is our model with predictions. Various ML model were applied and the results were compared.

    See project

  • Amazon-Food-Reviews-Analysis-and-Modelling

    The purpose of this analysis is to make up a prediction model where we will be able to predict whether a recommendation is positive or negative. In this analysis, we will not focus on the Score, but only the positive/negative sentiment of the recommendation. Various ML model were applied and the results were compared.

    See project

  • Optimization of Energy Consumption by Creating Uniformly Densed & Fixed Clusters in WSN

    -

    Our objective was to increase the life span of WSN and the problem statement was to generate uniform cluster of closer nodes. Therefore, by forming uniform cluster of closer nodes, we have got efficient results in terms of energy consumption by the sensor nodes for information interchange.. Firstly, we've developed algorithm for uniform clustering,then we've implemented the simulations on the MATLAB.

Languages

  • Hindi

    -

  • British English

    -

Recommendations received

1 person has recommended Abhishek

Join now to view

More activity by Abhishek

View Abhishek’s full profile

  • See who you know in common
  • Get introduced
  • Contact Abhishek directly
Join to view full profile

Other similar profiles

Explore more posts

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 More

Others named Abhishek Kumar in India

78273 others named Abhishek Kumar in India are on LinkedIn

See others named Abhishek Kumar

Add new skills with these courses

See all courses