Aspect Based Sentiment Analysis using GPT-3 | Prompt Design

Поделиться
HTML-код
  • Опубликовано: 10 янв 2025

Комментарии • 38

  • @FutureSmartAI
    @FutureSmartAI  Год назад +3

    📌 Hey everyone! Enjoying these NLP tutorials? Check out my other project, AI Demos, for quick 1-2 min AI tool demos! 🤖🚀
    🔗 RUclips: www.youtube.com/@aidemos.futuresmart
    We aim to educate and inform you about AI's incredible possibilities. Don't miss our AI Demos RUclips channel and website for amazing demos!
    🌐 AI Demos Website: www.aidemos.com/
    Subscribe to AI Demos and explore the future of AI with us!

  • @coolraviraj24
    @coolraviraj24 Год назад

    this video is really helpful it made me understand about the prompt

  • @efeokudan372
    @efeokudan372 2 года назад +1

    Great video. Actually golden for these days

  • @FutureSmartAI
    @FutureSmartAI  2 года назад +1

    I have added a link to the notion page, which Includes Instructions or Prompt used during Video

  • @mansibisht557
    @mansibisht557 Год назад +1

    Great :) Succinct and informative video. I look forward to seeing more of your content. :)

  • @sudhakarkaushik6128
    @sudhakarkaushik6128 2 года назад +1

    Loved the examples and incremental steps to get to the aspect. thanks Pradip

  • @prajithkumar432
    @prajithkumar432 2 года назад +1

    Kudos Pradip, very well explained👍👍

  • @yashjain6372
    @yashjain6372 Год назад +1

    Awesome video,learnt a lot.
    Question : What alternative of GPT-3 we could use to solve this problem from Open source LLM (Hugging face comerical LLM Licenced) ?

    • @FutureSmartAI
      @FutureSmartAI  Год назад

      Did you chak llama2 : huggingface.co/docs/transformers/model_doc/llama2

  • @s.sasisekhar4608
    @s.sasisekhar4608 2 года назад +1

    This is very good

    • @FutureSmartAI
      @FutureSmartAI  2 года назад +1

      Thank You. Keep watching any suggestions for next video?

    • @s.sasisekhar4608
      @s.sasisekhar4608 2 года назад

      @@FutureSmartAI Can you show multiple ways of doing EDA on text data (Granular Level)

    • @FutureSmartAI
      @FutureSmartAI  2 года назад

      @@s.sasisekhar4608 Sure in future videos. Currently still focusing on NLP use cases.

  • @AboniaSojasingarayar
    @AboniaSojasingarayar Год назад

    Hello Pradip , thanks for this valuable video.With aspect pair , is't possible to get the sentiment type of overall text as positive, negative ,neutral and mixed ? Thanks

  • @mrsamvs
    @mrsamvs Год назад +1

    Giving bad git executed error on local jupyter systems upon installing pyabsa. Any suggestions on how to fix it?

  • @zohaibramzan6381
    @zohaibramzan6381 2 года назад +1

    Kindly provide
    Link to video: Aspect base sentiment analysis using bert model

    • @FutureSmartAI
      @FutureSmartAI  2 года назад

      ruclips.net/video/-UEU-HGjUyQ/видео.html

  • @sachinbagmar1167
    @sachinbagmar1167 Год назад

    Hi Pradip thanks for sharing the prompt lists. How to go about classify bulk reviews? will the api call needs to be done for each review or is there a way to do bulk classifications of data. Video / response on this will be great ! thanks in advance for sharing this insights

    • @FutureSmartAI
      @FutureSmartAI  Год назад

      You can write prompt which takes list of review and returns list of sentiments.

  • @zohaibramzan6381
    @zohaibramzan6381 2 года назад +1

    Bro if we want to fine-tune the model, then how many prompt and completion pairs would be enough?? Secondly, what should be the structure of prompt and completion?

    • @FutureSmartAI
      @FutureSmartAI  2 года назад

      You can start with 20 example pairs and check performance if needed you can add more.
      Prompt Structure depends on what output you are expecting. Your sample examples should be in same format

  • @SheikhBelayetMahmood
    @SheikhBelayetMahmood Год назад

    I have one question. For example if the input is "The design of iphone is good but the samsung has longer battery life ". Now if I want to analyze the sentiment of iphone is clear that the sentiment about design is positive but for battery life is negative. How to design the prompt for this particular example. Since there are 2 brands are mentioned here.

    • @FutureSmartAI
      @FutureSmartAI  Год назад

      First you write down sample examples and expected output and GPT should be able to produce similar way

  • @rizwankhanniazi3620
    @rizwankhanniazi3620 Год назад

    you have done with the project of aspect sentiment analysis and upload on your upwork profile Kindle share you full project code of aspect sentimental anslysis with deployment

    • @FutureSmartAI
      @FutureSmartAI  Год назад

      It is similar, each video on youtube has code

  • @saralasri9129
    @saralasri9129 Год назад

    Hi Pradip, how can i solve this problem ? InvalidRequestError: The model `curie:ft-wrAQszDv88OVOWOQSjjqLZqe` does not exist

    • @FutureSmartAI
      @FutureSmartAI  Год назад

      It means model you are referring is not attched to your account. Is this valid model?

  • @afsheenmaroof6209
    @afsheenmaroof6209 Год назад

    hello how can i use this with my own data set?

    • @FutureSmartAI
      @FutureSmartAI  Год назад

      It will be simple juse use pandas library to read your dataset . iterate through it and call gpt for each row

    • @afsheenmaroof6209
      @afsheenmaroof6209 Год назад

      @FutureSmartAI can you elaborte this how to call...

  • @rudreshmehta6510
    @rudreshmehta6510 2 года назад

    Is there a way expect GPT3? Any small model?

    • @FutureSmartAI
      @FutureSmartAI  2 года назад

      Hi Rudresh, I couldn't get ? can you elaborate bit?