Update README.md

README.md

@@ -11,44 +11,27 @@ widget:
     example_title: Twitter user profile
 ---
 
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
-
-# twitter-health-users
-
-This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset.
-
 ## Model description
 
-Use this model to detect Twitter user's profiles related to healthcare.
+Use this model to detect Twitter users' profiles related to healthcare.
 
-Work in progress!
+User profile classification may be useful when searching for health information on Twitter. For a certain health topic, tweets from physicians or organizations may be more reliable than undefined profiles.
 
+The receives a user description text as input and returns a descriptive label for each profile:
 
-## Intended uses & limitations
+- `not-health-related`
+- `health-related`
+  - `health-related/person`
+  - `health-related/organization`
+  - `health-related/publishing`
+  - `health-related/physician`
+  - `health-related/news`
+  - `health-related/academic`
 
-More information needed
+This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on a dataset of user description texts that were semi-automatically labeled using regex parsing.
 
-## Training and evaluation data
+## Evaluation
 
 More information needed
 
-## Training procedure
-
-### Training hyperparameters
-
-The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 16
-- eval_batch_size: 16
-- seed: 42
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- num_epochs: 2
-
-### Framework versions
 
-- Transformers 4.26.0
-- Pytorch 1.13.1+cu116
-- Datasets 2.9.0
-- Tokenizers 0.13.2