Redoan Rahman

Redoan Rahman

Mountain View, California, United States
453 followers 450 connections

Join to view profile

About

I am a dedicated software and data engineer with a Master’s in Information Science from…

Experience

Education

Licenses & Certifications

Publications

  • Disparity in the Evolving COVID-19 Collaboration Network

    Springer Nature Switzerland

    COVID-19 pandemic has paused many ongoing research projects and unified researchers’ attention to focus on COVID-19 related issues. Our project traces 712,294 scientists’ publications related to COVID-19 for two years, from January 2020 to December 2021, in order to detect the dynamic evolution patterns of COVID-19 collaboration network over time. By studying the collaboration network of COVID-19 scientists,
    we observe how a new scientific community has been built in preparation for a sudden…

    COVID-19 pandemic has paused many ongoing research projects and unified researchers’ attention to focus on COVID-19 related issues. Our project traces 712,294 scientists’ publications related to COVID-19 for two years, from January 2020 to December 2021, in order to detect the dynamic evolution patterns of COVID-19 collaboration network over time. By studying the collaboration network of COVID-19 scientists,
    we observe how a new scientific community has been built in preparation for a sudden shock. The number of newcomers grows incrementally, and the connectivity of the collaboration network shifts from loose to tight promptly. Even though every scientist has an equal opportunity to start a study, collaboration disparity still exists. Following the scale-free distribution, only a few top authors are highly connected with other authors. These top authors are more likely to attract newcomers and work with each other. As the
    collaboration network evolves, the increase rate in the probability of attracting newcomers for authors with higher degree increases, whereas the increase rates in the probability of forming new links among authors with higher degree decreases. This highlights the interesting trend that COVID pandemic alters the research collaboration trends that star scientists are starting to collaborate more with newcomers, but less with existing collaborators, which, in certain way, reduces the collaboration disparity.

    Other authors
    See publication

  • Using Explainable AI to Cross-Validate Socio-economic Disparities Among Covid-19 Patient Mortality

    AMIA 2023 Informatics Summit

    This paper applies eXplainable Artificial Intelligence (XAI) methods to investigate the socioeconomic disparities in COVID patient mortality. An Extreme Gradient Boosting (XGBoost) prediction model is built based on a de-identified Austin area hospital dataset to predict the mortality of COVID-19 patients. We apply two XAI methods, Shapley Additive exPlanations (SHAP) and Locally Interpretable Model Agnostic Explanations (LIME), to compare the global and local interpretation of feature…

    This paper applies eXplainable Artificial Intelligence (XAI) methods to investigate the socioeconomic disparities in COVID patient mortality. An Extreme Gradient Boosting (XGBoost) prediction model is built based on a de-identified Austin area hospital dataset to predict the mortality of COVID-19 patients. We apply two XAI methods, Shapley Additive exPlanations (SHAP) and Locally Interpretable Model Agnostic Explanations (LIME), to compare the global and local interpretation of feature importance. This paper demonstrates the advantages of using XAI which shows the feature importance and decisive capability. Furthermore, we use the XAI methods to cross-validate their interpretations for individual patients. The XAI models reveal that Medicare financial class, older age, and gender have high impact on the mortality prediction. We find that LIME local interpretation does not show significant differences in feature importance comparing to SHAP, which suggests pattern confirmation. This paper demonstrates the importance of XAI methods in cross-validation of feature attributions.

    Other authors
    See publication

  • Analyzing Impact of Socio-Economic Factors on COVID-19 Mortality Prediction Using SHAP Value

    AMIA 2022 Annual Symposium

    This paper applies multiple machine learning (ML) algorithms to a dataset of de-identified COVID-19 patients provided by the COVID-19 Research Database. The dataset consists of 20,878 COVID-positive patients, among which 9,177 patients died in the year 2020. This paper aims to understand and interpret the association of socio-economic characteristics of patients with their mortality instead of maximizing prediction accuracy. According to our analysis, a patients households annual and disposable…

    This paper applies multiple machine learning (ML) algorithms to a dataset of de-identified COVID-19 patients provided by the COVID-19 Research Database. The dataset consists of 20,878 COVID-positive patients, among which 9,177 patients died in the year 2020. This paper aims to understand and interpret the association of socio-economic characteristics of patients with their mortality instead of maximizing prediction accuracy. According to our analysis, a patients households annual and disposable income, age, education, and employment status significantly impacts a machine learning models prediction. We also observe several individual patient data, which gives us insight into how the feature values impact the prediction for that data point. This paper analyzes the global and local interpretation of machine learning models on socio-economic data of COVID patients.

    Other authors
    See publication

  • A Hierarchical Learning Model for Claim Validation

    Springer

    Due to the proliferation of social media platforms, people increasingly depend on these platforms to consume news content. Consequently, propagandists utilize these platforms to easily spread fake or distorted contents. Hence, the fake news detection has been a crucial but difficult task so far, due to the lack of comprehensive labeled datasets and the diverse linguistics cues in the fake news statements. However, recently to aid the designing of a computational model for validating the news…

    Due to the proliferation of social media platforms, people increasingly depend on these platforms to consume news content. Consequently, propagandists utilize these platforms to easily spread fake or distorted contents. Hence, the fake news detection has been a crucial but difficult task so far, due to the lack of comprehensive labeled datasets and the diverse linguistics cues in the fake news statements. However, recently to aid the designing of a computational model for validating the news contents, a few labeled benchmark datasets have been introduced in the literature, such as LIAR dataset. In this work, we augment the LIAR dataset’s claim statements and the speakers’ profile features with the evidence retrieved from the Politifact. We utilize this augmented dataset to design a transfer learning-based claim verification model, TLCV: Transfer Learning based Claim Validation. Moreover, in TLCV, we design an evidence retrieval module to extract the appropriate evidence. We embed these feature groups by leveraging the pretrained ELMo model, which enables TLCV model to capture the deep contextual feature representation. We have designed a hierarchical Convolutional Neural Network to create a composite feature representation. In the performance analysis, it is shown that our proposed claim verification approach, TLCV, outperforms the baseline approach by 8.81% on LIAR dataset.

    Other authors
    See publication

Projects

  • Articulating Data Quality for Effective Space Situational Awareness

    -

    This research works as a stepping stone for developing a data pipeline allowing for effective space object tracking and improving space situational awareness. In this research, we develop a framework to analyze the quality of the received data from various sources, define performance metrics to measure the quality of the received data and determine the overlap of observed objects between the involved sources.

    Other creators

  • Dining at the University of Texas(Redesigning UT Dining website)

    -

    The UT Dining website page is an information resource for UT students, faculty, staff, and visiting family members. The website aims to provide a reliable and easy-to-use UT dining information experience. In this project, we aim to redesign the website to enable users to discover and experience UT dining by providing a clear visualization of information & the ability to find eating places according to their needs efficiently. Having a reliable information source will enable users to plan &…

    The UT Dining website page is an information resource for UT students, faculty, staff, and visiting family members. The website aims to provide a reliable and easy-to-use UT dining information experience. In this project, we aim to redesign the website to enable users to discover and experience UT dining by providing a clear visualization of information & the ability to find eating places according to their needs efficiently. Having a reliable information source will enable users to plan & execute their dining plans & schedules comfortably while making them feel like a part of the larger UT community.

    Other creators

  • Pandemic Simulation

    -

    In this project, we set a fixed population and an infection rate and tried to determine the time it takes to achieve herd immunity. We added modifiers such as vaccination rate, random chance of infection, et cetera.

  • COVID-19 mortality prediction

    -

    There have been many types of research on predicting mortality due to Coid-19 based on medical data. However, these data are collected after the patient has been infected; by then, the patient is already in danger. Different researches prove that the socio-economic characteristics of a human being can help identify their susceptibility to certain diseases. In this project, we partnered with Bill & Melinda Gates Foundation to find the significance of the socio-economic characteristics of people…

    There have been many types of research on predicting mortality due to Coid-19 based on medical data. However, these data are collected after the patient has been infected; by then, the patient is already in danger. Different researches prove that the socio-economic characteristics of a human being can help identify their susceptibility to certain diseases. In this project, we partnered with Bill & Melinda Gates Foundation to find the significance of the socio-economic characteristics of people to predict the danger. We use different ML models to predict patient death due to Covid-19. Then we analyze the model’s feature importance through different explainable AI methods and determine which socio-economic characteristics are more indicative of Covid-19 risks.

    Other creators

  • Visual Graph Query Builder

    -

    The project involved developing a custom graph query builder system for users to visualize the changes made to the query dynamically. The system's output was fed directly into a graph query processor, allowing users to connect the result with their search parameters more seamlessly.

    Other creators

  • Relation Extraction

    -

    The research project explores different possibilities to build a successful model that can classify the type of relation type available in a sentence of medical text. Our research utilized various feature extraction methods such as TF-IDF, Bag-of-Words, Word2Vec, Spacy, BERT, and Sentence-BERT. For the classification task, we implemented models such as Decision Tree, Random Forest, Long Short Term Memory model, et cetera. We found that in our smaller datasets, simpler models do surprisingly…

    The research project explores different possibilities to build a successful model that can classify the type of relation type available in a sentence of medical text. Our research utilized various feature extraction methods such as TF-IDF, Bag-of-Words, Word2Vec, Spacy, BERT, and Sentence-BERT. For the classification task, we implemented models such as Decision Tree, Random Forest, Long Short Term Memory model, et cetera. We found that in our smaller datasets, simpler models do surprisingly better than the complex model. We also found that SentenceBERT provides excellent representation, improving the results for almost all classification models. We will use our findings to advance the research further.

    Other creators

  • Computational Approach for Deceptive Content Detection

    -

    The essence of the thesis is to develop an algorithm to detect fake news by distinguishing between fabricated, misleading information and correct information. For this purpose, a large data set containing a considerable number of claims, verdicts, and statements have been used to train and test the model. Recurrent Neural Network models such as Long Short Term Memory Model(LSTM) and Gated Recurrent Unit(GRU) and feature representation mechanisms such as Sent2Vec, ELMo, and ULMFit have also been…

    The essence of the thesis is to develop an algorithm to detect fake news by distinguishing between fabricated, misleading information and correct information. For this purpose, a large data set containing a considerable number of claims, verdicts, and statements have been used to train and test the model. Recurrent Neural Network models such as Long Short Term Memory Model(LSTM) and Gated Recurrent Unit(GRU) and feature representation mechanisms such as Sent2Vec, ELMo, and ULMFit have also been utilized to create the model for detecting fake news.

    Other creators

  • Backward Propagation

    -

    A functioning android application developed using android studio that educates the users about the process of learning of a basic neural network by showing the visualization of the learning process of different binary operations
    • Backend: Python (Django Framework), Frontend: HTML, CSS, Javascript
    • Responsibilities: Frontend, Testing, Backend (Partial)

  • BoiKoi

    -

    A social website dedicated to providing a community comprising of bookworms and scholars where people can find enlightened intellectuals with shared literature interests.
    • Used Django framework.
    • Backend: Python
    • Frontend: HTML, CSS, Javascript

  • Genre Detection of Books

    -

    A project to determine the probability of a book belonging to a genre from the plot summary using word2vec and spherical K-means clustering.
    • Programming language: Python
    • Responsibilities: Data Collection and Implementation (Partial)

  • VizWiz

    -

    A survey tool developed with the sole purpose of making the process of survey creation and data collection simpler. It also provided data representation functionality to make an analyzation of survey results easier as well as a web portal providing access to survey data which also contained data visualization functionality.
    • Patterns Used: Strategy pattern, Template method pattern, Null Object pattern, Iterator pattern, Singleton pattern, Decorator pattern, Observer pattern.
    • Used…

    A survey tool developed with the sole purpose of making the process of survey creation and data collection simpler. It also provided data representation functionality to make an analyzation of survey results easier as well as a web portal providing access to survey data which also contained data visualization functionality.
    • Patterns Used: Strategy pattern, Template method pattern, Null Object pattern, Iterator pattern, Singleton pattern, Decorator pattern, Observer pattern.
    • Used Android Studio for the survey tool development
    • Used Django for Web Portal
    • Used Firebase for Database service

    Other creators
    See project

  • Virtual Glove

    -

    A peripheral device created using a simple glove, flex registers and Arduino that converts hand gestures into machine-readable information that allows interaction with 3D objects created in the Unity environment.
    • Used Unity for 3D Modeling and Environment creation
    • Used Arduino for Hardware Interfacing

    Other creators

  • Goalkeeper Performance in European Leagues

    -

    A complete analysis of the performance of goalkeepers in European Football Leagues based on their clean sheets, goals consumed and other available data
    • Responsibilities: Data Collection, Analysis and Documentation

  • ChildCare

    -

    A functioning application built with Android studio that educates expecting mothers about childcare during the time of pregnancy as well as when the baby is newborn.
    • Developed using Android Studio

    Other creators

  • Dragon Ball Z: Frieza Saga

    -

    A simple 2D game that allows the player to fight against a basic artificial intelligence
    • Programming language: C/C++, Borland Graphics Interface(BGI)
    • Responsibilities: Development and Testing

    Other creators

Languages

  • Bengali

    Native or bilingual proficiency

  • English

    Professional working proficiency

Recommendations received

6 people have recommended Redoan

Join now to view

View Redoan’s full profile

  • See who you know in common
  • Get introduced
  • Contact Redoan 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 Redoan Rahman

3 others named Redoan Rahman are on LinkedIn

See others named Redoan Rahman

Add new skills with these courses

See all courses