This program tweets a pie chart that shows what the user thinks about the keyword by inputting the keyword you wanna search into the shortcuts app (provided by Apple).
After a text file containing a keyword is posted to a specific folder in Dropbox, the PC starts collecting the tweet data by downloading it.
You can select whether you wanna search with only tweet text or text + image of the tweet.
Prepare on the assumption that you already have TWITTER & DROPBOX API Key.
-
Install packages you need to do to run the program.
-
Download shortcuts recipe GLOBAL PBL on your iPhone or iPad series.
-
Input your twitter API keys to api.py.
-
Input your DROPBOX API Key to main.py.
-
Input "PBL-Everyone-" folder location on your PC to PBL_FILE_PATH of const.py.
-
Run
python main.py -
Run the shortcut GLOBAL PBL and input search key word and answer yes or no question.
-
Keep watching over it...(Just a few second)
-
Check your tweet and you can know what you wanna know.
This program outputs each value of Positive, Neutral and Negative.
But the image processing, which is a part of this program, returns the values of Angry, Disgust, Fear, Sad, Surprise, Neutral and Happy.
So There is a need for converging them to those values.
For that reason, the above processing is performed.
However, this is not an academic calculation method.
If you know the right calculation way, please tell me.
Copyright (c) 2020 Yusuke Sakabe
Released under the MIT license
Provided and inspirated by
https://github.com/oarriaga/face_classification
https://github.com/petercunha/Emotion
https://github.com/dropbox/dropbox-sdk-python