数据集获取

git clone https://github.com/abuBakarSiddiqurRahman/DepressionEmo.git

curl -L -o repo.zip https://github.com/abuBakarSiddiqurRahman/DepressionEmo/archive/refs/heads/main.zip
unzip repo.zip

源站 README 摘录（使用方式）

DepressionEmo

“DepressionEmo: A novel dataset for multilabel classification of depression emotions”
https://www.sciencedirect.com/science/article/abs/pii/S0165032724012278
https://arxiv.org/pdf/2401.04655.pdf
If you use our dataset in your research paper, please make sure to cite the following paper:
Rahman, A. B. S., Ta, H. T., Najjar, L., Azadmanesh, A., & Gönül, A. S. (2024). DepressionEmo: A novel dataset for multilabel classification of depression emotions. Journal of Affective Disorders. https://doi.org/10.1016/j.jad.2024.08.013
Use of the DepressionEmo dataset or a part of your whole dataset for developing any web application, mobile app, or commercial tool is strictly prohibited without prior written permission.
Please contact us at abu.rahman@hamptonu.edu before initiating any such use.
© 2024 Abu Bakar Siddiqur Rahman. All Rights Reserved.

Dataset

Subsets

The dataset is divided into 3 subsets:

• Training set
• Validation set
• Test set

An example data

Each example contains “id”, “title”, “post”, “text”, “upvotes”, “date”, “emotions”, and “label_id”. We use “text” (concatenate from “title” and “post”) for the depression detection.

{"id": "zq1lwl", "title": "goodbye.", "post": "I''m done. I have a bottle of jack danials and couple bottles of sleeping meds waiting for me when I get home. \n\nI''m excited to leave this place. There will be no regret, no loneliness, no sorrow, and no misery. I am looking forward to the peace of nonexistence.\n\nI love you all.", "text": "goodbye. ### I''m done. I have a bottle of jack danials and couple bottles of sleeping meds waiting for me when I get home. I''m excited to leave this place. There will be no regret, no loneliness, no sorrow, and no misery. I am looking forward to the peace of nonexistence. I love you all.", "upvotes": 102, "date": "2022-12-19 19:50:52", "emotions": ["emptiness", "hopelessness"], "label_id": 110000}

There are 8 depression emotions:

emotionevent-blocked= [''anger'', ''brain dysfunction (forget)'', ''emptiness'', ''hopelessness'', ''loneliness'', ''sadness'', ''suicide intent'', ''worthlessness'']

The “label_id” field is a number, whose each digit represents a state of with emotion (1) and without emotion (0). For example:

• [“emptiness”, “hopelessness”] -> 00110000 -> 110000
• [“anger”] -> 10000000
• [‘‘anger’’, ‘‘brain dysfunction (forget)’’, ‘‘emptiness’’, ‘‘hopelessness’’, ‘‘loneliness’’, ''sad

数据加载示例（表格/文本类）

import pandas as pd, glob, os

files = (glob.glob(os.path.join(path, "**", "*.csv"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.tsv"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.xlsx"), recursive=True))
print("数据文件:", files)
df = pd.read_csv(files[0])
print(df.shape); print(df.columns.tolist()); print(df.head(3))

完整仓库：github.com/abuBakarSiddiqurRahman/DepressionEmo