跳转到主要内容

实用深度学习和机器学习项目创意的精选列表

  • 30 多个想法
  • 与学术界和工业界相关
  • 范围从初学者友好到研究项目

Contents

  • Text - With some topics about Natural language processing

  • Forecasting - Most of the topics in this section is about Time Series and similar forecasting challenges

  • Recommendation Systems

  • Vision - With topics about image and video processing

  • Covid19 - Multi or Single Domain ideas from the Covid19 theme

  • Music and Audio - These topics are about combining ideas from language and audio to understand music

  • Conclusion


Covid19

  • Bing Coronavirus

    • Classify Bing Queries as either specific (e.g. about a specific location) or generic. You might have to figure out a more exact definition of specific or generic though
    • Dataset: BingCoronavirusQuerySet
  • Covid Clinical Data

    • Rank and sort high risk patients using clinical data. Pick an interpretable approach if you can.
    • Dataset: CovidClinicalData

If you haven't already, checkout Kaggle's Covid19 Section as well. It has datasets and ideas both.

Text

Natural Language Understanding

  • Sentence to Sentence semantic similarity

    • Can you identify question pairs that have the same intent or meaning?
    • Dataset: Quora question pairs with similar questions marked
  • Fight online abuse

  • Open Domain question answering

  • Automatic text summarization

    • Can you create a summary with the major points of the original document?
    • Abstractive (write your own summary) and Extractive (select pieces of text from original) are two popular approaches
    • Dataset: CNN and DailyMail News Pieces by Google DeepMind
  • Copy-cat Bot

Check mlm/blog for some hints.

Forecasting

  • Univariate Time Series Forecasting

  • Multi-variate Time Series Forecasting

    • How polluted will your town's air be? Pollution Level Forecasting
    • Dataset: Air Quality dataset
  • Demand/load forecasting

  • Predict Blood Donation

    • We're interested in predicting if a blood donor will donate within a given time window.
    • More on the problem statement at Driven Data.
    • Dataset: UCI ML Datasets Repo

Recommendation systems

  • Movie Recommender

    • Can you predict the rating a user will give on a movie?
    • Do this using the movies that user has rated in the past, as well as the ratings similar users have given similar movies.
    • Dataset: Netflix Prize and MovieLens Datasets
  • Search + Recommendation System

    • Predict which Xbox game a visitor will be most interested in based on their search query
    • Dataset: BestBuy
  • Can you predict Influencers in the Social Network?

    • How can you predict social influencers?
    • Dataset: PeerIndex

Vision

  • Image classification

  • Bone X-Ray competition

    • Can you identify if a hand is broken from a X-ray radiographs automatically with better than human performance?
    • Stanford's Bone XRay Deep Learning Competition with MURA Dataset
  • Image Captioning

    • Can you caption/explain the photo a way human would?
    • Dataset: MS COCO
  • Image Segmentation/Object Detection

  • Large-Scale Video Understanding

    • Can you produce the best video tag predictions?
    • Dataset: YouTube 8M
  • Video Summarization

  • Style Transfer

    • Can you recompose images in the style of other images?
    • Dataset: fzliu on GitHub shared target and source images with results
  • Chest XRay

    • Can you detect if someone is sick from their chest XRay? Or guess their radiology report?
    • Dataset: MIMIC-CXR at Physionet
  • Clinical Diagnostics: Image Identification, classification & segmentation

    • Can you help build an open source software for lung cancer detection to help radiologists?
    • Link: Concept to clinic challenge on DrivenData
  • Satellite Imagery Processing for Socioeconomic Analysis

    • Can you estimate the standard of living or energy consumption of a place from night time satellite imagery?
    • Reference for Project details: Stanford Poverty Estimation Project
  • Satellite Imagery Processing for Automated Tagging

    • Can you automatically tag satellite images with human features such as buildings, roads, waterways and so on?
    • Help free the manual effort in tagging satellite imagery: Kaggle Dataset by DSTL, UK

Music

  • Music/Audio Recommendation Systems

  • Music Genre recognition using neural networks

    • Can you identify the musical genre using their spectrograms or other sound information?
    • Datasets: FMA or GTZAN on Keras
    • Get started with Librosa for feature extraction

原文:https://github.com/NirantK/awesome-project-ideas