README.md

---
license: mit
tags:
- multilingual
- text
- coordinates
- geospatial
- translation
- NER
- geo
- geo-tagged
- named-entity-recognition
- natural-language-processing
- geographic-data
- geolocation
- twitter
- reddit
task_categories:
- feature-extraction
- token-classification
- text-classification
pretty_name: Geo-Tagged Social Media Posts with Timestamps
language:
- en
- es
- ru
- co
- ar
- fa
size_categories:
- 100M<n<1B
---

# Dataset Card for Geo-Tagged Social Media Posts with Timestamps

## Dataset Description

- **Homepage:**  https://huggingface.co/datasets/yachay/text_coordinates_seasons
- **Repository:** https://github.com/Yachay-AI/byt5-geotagging#datasets
- **Paper:** https://dev.to/yachayai/applying-machine-learning-to-geolocate-twitter-posts-2m1d
- **Leaderboard:** 
- **Point of Contact:** [email protected]

### Dataset Summary

The "Seasons" dataset is a collection of over 600,000 social media posts spanning 12 months and encompassing 15 distinct time zones. It focuses on six countries: **Cuba, Iran, Russia, North Korea, Syria, and Venezuela,** with each post containing textual content, timestamps, and geographical coordinates. The dataset's primary objective is to investigate the correlation between the timing of posts, their content, and the geographical locations. Researchers can leverage this dataset to advance studies in geospatial NLP and gain insights into how temporal factors and seasonality impact the results.

### Supported Tasks and Leaderboards

This dataset is well-suited for tasks such as geotagging, where the objective is to associate text with specific geographical locations. It can also be utilized for geolocation analysis, sentiment analysis in regional contexts, and regional text classification.

### Languages

**Multilingual Dataset**

Mainly contains English, Spanish, Persian, Russian, Korean, and Arabic.

## Dataset Structure

### Data Instances

The "Seasons" dataset consists of over 600,000 data instances, each representing a social media post. 

### Data Fields

**Text (text):** This field contains the textual content.

**Timestamp (created_at):** The dataset includes timestamps to track the exact time when each social media post was created. Timestamps are recorded in Unix epoch time format.

**Geographical Coordinates (geo_geo_bbox):** This field contains geocoordinates that describe the geographical location associated with each social media post. These coordinates are represented as latitude and longitude ranges in a bounding box format.

```json

{
  "text": "sample text",
  "geo_geo_bbox": "[-67.220209, 9.934294, -65.428322, 10.6496277]"
},
{
  "created_at": {
    "$numberLong": "1633049378000"
  }

```

### Data Splits

This dataset is not pre-partitioned into training, validation, and test data splits, providing flexibility for users to split the data according to their specific research or application needs. Users can customize the data partitioning to suit their machine learning experiments and analytical requirements.

## Dataset Creation

### Curation Rationale

The "Seasons" dataset was created with an objective to advancing research in NLP by investigating the intricate relationships between temporal factors, content, and author location in social media posts. This dataset was assembled to provide a resource for understanding how time zones and seasonal events influence the model's results. 

### Source Data

#### Initial Data Collection and Normalization

The initial data collection process focused on gathering geotagged comments from social media platforms, with a primary emphasis on Twitter.

#### Who are the source language producers?

Twitter Community 

### Annotations

#### Annotation process

The coordinates in this dataset have been derived from metadata sources.

#### Who are the annotators?

No manual annotation was conducted for this dataset.

## Considerations for Using the Data

### Social Impact of Dataset

The "Seasons" dataset has a potential to enhance our understanding of the intricate relationship between temporal dynamics, content, and location in social media posts.

### Discussion of Biases

It's essential to acknowledge that the data collected from social media platforms may contain inherent biases, influenced by user demographics and platform dynamics. Researchers should be mindful of these biases and consider potential implications in their analyses.

### Other Known Limitations

- The dataset's multilingual nature may lead to varying data quality and linguistic diversity across regions.
- The use of geotagged social media comments means that the dataset may not cover less active or less represented regions/seasons. 
- The accuracy of geocoordinates is subject to inherent limitations of the data sources used for collection.
  
## Additional Information

### Dataset Curators

Yachay AI

### Licensing Information

MIT