Scribd Logo
Upload

Welcome to Scribd!

  • 188 views

    Data Science Problem Statements

    Uploaded by

    Rishav Choudhary
    This document outlines three data science problem statements: 1) Build an intelligent corporate joker application using NLP that can respond to jokes based on input or without input. 2) Create a video summarization model that can generate summaries of uploaded videos and allow searching videos by text. 3) Develop a reverse image search engine that can find database images matching a query image. All three problems require data collection, model training and deployment through a web application while following MLOPS practices.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Data Science Problem Statements

    Uploaded by

    Rishav Choudhary
    188 views3 pages
    This document outlines three data science problem statements: 1) Build an intelligent corporate joker application using NLP that can respond to jokes based on input or without input. 2) Create a video summarization model that can generate summaries of uploaded videos and allow searching videos by text. 3) Develop a reverse image search engine that can find database images matching a query image. All three problems require data collection, model training and deployment through a web application while following MLOPS practices.

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document outlines three data science problem statements: 1) Build an intelligent corporate joker application using NLP that can respond to jokes based on input or without input. 2) Create a video summarization model that can generate summaries of uploaded videos and allow searching videos by text. 3) Develop a reverse image search engine that can find database images matching a query image. All three problems require data collection, model training and deployment through a web application while following MLOPS practices.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    188 views3 pages

    Data Science Problem Statements

    Uploaded by

    Rishav Choudhary
    This document outlines three data science problem statements: 1) Build an intelligent corporate joker application using NLP that can respond to jokes based on input or without input. 2) Create a video summarization model that can generate summaries of uploaded videos and allow searching videos by text. 3) Develop a reverse image search engine that can find database images matching a query image. All three problems require data collection, model training and deployment through a web application while following MLOPS practices.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 3

    Data Science Problem Statements

    Problem Statement 1 (Corporate Joker)

    Build an Intelligent corporate joker application using NLP. The model should reply with jokes as
    demanded based on circumstances.

    Domain: NLP

    How to collect data?


    Tips: Use joke websites from the internet and scrap data from there.

    Requirement:
    1. Create a function that can return jokes without input.
    2. Create a function that can accept a joke and respond joke based on the joke provided.
    Example:
    Input: How did pirates communicate before the internet?
    Response: Pier to Pier Networking.

    What to Submit:
    ● Data collection?
    ● Steps of data wrangling.
    ● Model Training.
    ● Model Evaluation.
    ● Model Deployment.
    ● Create a web-based application to demonstrate the build solution.
    ● Follow MLOPS practices to build the project.

    Note: Demonstrate the provided requirement using a web application.


    Problem Statement 2 (Video Summarization)

    Description:
    Build the video summarization model to extract a summary of a video. Summarized video content
    helps us to understand the video content without watching a video. Use summarised text to build a
    video search engine, which will help to search a video just by the video’s description.

    Tip: Use NLP techniques such as embedding to build a text-based search engine.

    How to collect data?


    Collect data using online resources.

    Requirement:
    1. Generate a summary of the uploaded video.
    2. Searching videos using text.

    What to Submit:
    ● Data collection?
    ● Steps of data wrangling.
    ● Model Training.
    ● Model Evaluation.
    ● Model Deployment.
    ● Create a web-based application to demonstrate the solution built.
    ● Follow MLOPS practices to build the project.

    Note: Demonstrate the provided requirement using a web application.


    Problem Statement 3 (Reverse Image Search Engine)

    The built models should be able to find database images that correspond to a given query image
    (i.e., the model should retrieve database images containing the same object as the query).

    How to collect data?


    Collect data using online resources.

    Requirement:
    1. Image Search Engine.

    What to Submit:
    ● Data collection?
    ● Steps of data wrangling.
    ● Model Training.
    ● Model Evaluation.
    ● Model Deployment.
    ● Create a web-based application to demonstrate the build solution.
    ● Follow MLOPS practices to build the project.

    Note: Demonstrate the provided requirement using a web application.

    You might also like