language:
- en
tags:
- roberta
- marketing mix
- multi-label
- classification
- microblog
- tweets
widget:
- text: >-
Best cushioning ever!!! 🤗🤗🤗 my zoom vomeros are the bomb🏃🏽♀️💨!!!
@nike #run #training
- text: >-
Why is @BestBuy always sold-out of Apple's new airpods in their online
shop 🤯😡?
- text: They’re closing the @Aldo at the Lehigh Vally Mall and KOP 😭
- text: >-
@Sony’s XM3’s ain’t as sweet as my bro’s airpod pros but got a real steal
🤑 the other day #deal #headphonez
- text: >-
Nike needs to sponsor more e-sports atheletes with Air Jordans! #nike
#esports
- text: >-
Say what you want about @Abercrombie's 90s shirtless males ads, they made
dang good woll sweaters back in the day. This is one of 3 I have from the
late 90s.
- text: >-
To celebrate this New Year, @Nordstrom is DOUBLING all donations up to
$25,000! 🎉 Your donation will help us answer 2X the calls, texts, and
chats that come in, and allow us to train 2X more volunteers!
- text: >-
It's inspiring to see religious leaders speaking up for workers' rights
and fair wages. Every voice matters in the #FightFor15! 💪🏽✊🏼
#Solidarity #WorkersRights
Model Card for: mmx_classifier_microblog_ENv02
Multi-label classifier that identifies which marketing mix variable(s) a microblog post pertains to.
Version: 0.2 from August 16, 2023
Model Details
You can use this classifier to determine which of the 4P's of marketing, also known as marketing mix variables, a microblog post (e.g., Tweet) pertains to:
- Product
- Place
- Price
- Promotion
Model Description
This classifier is a fine-tuned checkpoint of [cardiffnlp/twitter-roberta-large-2022-154m] (https://huggingface.co./cardiffnlp/twitter-roberta-large-2022-154m). It was trained on 15K Tweets that mentioned at least one of 699 brands. The Tweets were first cleaned and then labeled using OpenAI's GPT4.
Because this is a multi-label classification problem, we use binary cross-entropy (BCE) with logits loss for the fine-tuning. We basically combine a sigmoid layer with BCELoss in a single class. To obtain the probabilities for each label (i.e., marketing mix variable), you need to "push" the predictions through a sigmoid function. This is already done in the accompanying python notebook.
IMPORTANT: At the time of writing this description, Huggingface's pipeline did not support multi-label classifiers.
Citation
For attribution, please cite the following reference if you use this model:
Ringel, Daniel, Creating Synthetic Experts with Generative Artificial Intelligence (July 15, 2023). Available at SSRN: https://ssrn.com/abstract=4542949
Working Paper
Download the working paper from SSRN: "Creating Synthetic Experts with Generative AI"