File size: 3,100 Bytes
b33da30
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0c2a059
 
 
b33da30
0c2a059
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d9a0f02
0c2a059
 
 
 
 
 
 
 
 
d9a0f02
0c2a059
 
 
 
 
d9a0f02
0c2a059
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: id
    dtype: string
  - name: user_id
    dtype: float64
  - name: date
    dtype: string
  - name: likes
    dtype: float64
  - name: shares
    dtype: int64
  - name: comments
    dtype: int64
  - name: post_text
    dtype: string
  - name: post_link
    dtype: string
  - name: img_link
    dtype: string
  - name: ocr
    dtype: string
  - name: caption
    dtype: string
  - name: reactions
    dtype: string
  splits:
  - name: train
    num_bytes: 10804037551.84
    num_examples: 235880
  download_size: 8561366079
  dataset_size: 10804037551.84
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
pretty_name: facebook_memes
size_categories:
- 100K<n<1M
---

# Dataset Card for Facebook Image Dataset

## Table of Contents
- [Dataset Description](#dataset-description)
  - [Dataset Summary](#dataset-summary)
  - [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
  - [Languages](#languages)
- [Dataset Structure](#dataset-structure)
  - [Data Instances](#data-instances)
  - [Data Fields](#data-fields)
  - [Data Splits](#data-splits)

## Dataset Description

### Dataset Summary
This dataset contains images scraped from Facebook along with associated metadata. The dataset is intended for research purposes, focusing on image analysis, natural language processing, and social media dynamics studies.

### Supported Tasks and Leaderboards
The dataset can be used for tasks such as image recognition, sentiment analysis, text extraction from images (OCR), and social media trend analysis.

### Languages
The text in the dataset is primarily in English, as extracted from Facebook posts.

## Dataset Structure

### Data Instances
A data instance consists of an image and its metadata. For example:
```json
{
  "id": "12345",
  "user_id": "67890",
  "date": "2023-07-04",
  "likes": 150,
  "shares": 25,
  "comments": 40,
  "post_text": "Here's a great moment captured #fun",
  "post_link": "https://facebook.com/example/status/12345",
  "img_link": "https://example.com/img.jpg",
  "ocr": "Here's a great moment captured",
  "captions": "A boy holding a balloon.",
  "reactions": ['Haha: 3 people', 'Love: 1 person']
  "file_name": "12345.jpg"
}
```
### Data Fields
- `id`: Unique identifier for each post.
- `user_id`: Facebook user ID of the post author.
- `date`: Date the post was made.
- `likes`: Number of likes the post received.
- `shares`: Number of shares the post received.
- `comments`: Number of comments on the post.
- `post_text`: Text content of the post.
- `post_link`: URL to the original Facebook post.
- `img_link`: URL to the image.
- `ocr`: Text extracted from the image using OCR.
- `captions`: An image description generated by [BLIP](https://github.com/salesforce/BLIP).
- `reactions`: Reactions native to Facebook, given by users for the post.
- `file_name`: Name of the file stored locally.

### Data Splits
This dataset does not have predefined splits (train/test/validation). Users can create splits as needed for their specific tasks.