SocialED.dataset

Dataloader

class SocialED.dataset.dataloader.DatasetLoader(dataset=None, dir_path=None)[source]

Bases: object

Base class for loading social event detection datasets.

Note

This is the base dataset loader class that provides common functionality for loading and preprocessing social event detection datasets. All specific dataset loaders should inherit from this class.

Parameters:
  • dataset (str, optional) – Name of the dataset to load. Default: None.

  • dir_path (str, optional) – Custom directory path to load data from. Default: None.

required_columns

Required columns that must be present in loaded datasets.

Type:

list

repo_url

URL of the repository containing the datasets.

Type:

str

target_folder

Target folder name for downloaded data.

Type:

str

download_and_cleanup(repo_url, dataset_name, local_target_folder)[source]
download()[source]
load_data()[source]

Temporary implementation that returns empty dataset

get_dataset_language()[source]

Determine the language based on the current dataset.

Returns:

The language of the dataset (‘English’, ‘French’, ‘Arabic’).

Return type:

str

get_dataset_name()[source]

Get the name of the current dataset.

Returns:

The name of the dataset.

Return type:

str

get_dataset_info()[source]

Get the info of the current dataset.

Returns:

The info of the dataset.

Return type:

list

class SocialED.dataset.dataloader.MAVEN(dir_path=None)[source]

Bases: DatasetLoader

The MAVEN dataset for social event detection.

Note

This dataset contains English language social media posts related to various events. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.CrisisNLP(dir_path=None)[source]

Bases: DatasetLoader

The CrisisNLP dataset for social event detection.

Note

This dataset contains English language social media posts related to crisis events. The dataset provides text content and event labels for crisis event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.Event2012(dir_path=None)[source]

Bases: DatasetLoader

The Event2012 dataset for social event detection.

Note

This dataset contains English language social media posts from 2012. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.Event2018(dir_path=None)[source]

Bases: DatasetLoader

The Event2018 dataset for social event detection.

Note

This dataset contains French language social media posts from 2018. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.Arabic_Twitter(dir_path=None)[source]

Bases: DatasetLoader

The Arabic Twitter dataset for social event detection.

Note

This dataset contains Arabic language tweets related to various events. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.CrisisLexT26(dir_path=None)[source]

Bases: DatasetLoader

The CrisisLexT26 dataset for social event detection.

Note

This dataset contains tweets related to 26 different crisis events. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.CrisisMMD(dir_path=None)[source]

Bases: DatasetLoader

The CrisisMMD dataset for social event detection.

Note

This dataset contains multimodal crisis-related social media data. The dataset provides text, images and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.HumAID(dir_path=None)[source]

Bases: DatasetLoader

The HumAID dataset for social event detection.

Note

This dataset contains tweets related to humanitarian crises and disasters. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.KBP(dir_path=None)[source]

Bases: DatasetLoader

The KBP dataset for social event detection.

Note

This dataset contains knowledge base population event data. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.Arabic_7(dir_path=None)[source]

Bases: DatasetLoader

The Arabic_7 dataset for social event detection.

Note

This dataset contains Arabic language social media posts for 7 event types. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.Event2012_100(dir_path=None)[source]

Bases: DatasetLoader

The Event2012_100 dataset for social event detection.

Note

This dataset contains tweets from 2012 related to 100 different events. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.Event2018_100(dir_path=None)[source]

Bases: DatasetLoader

The Event2018_100 dataset for social event detection.

Note

This dataset contains tweets from 2018 related to 100 different events. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.Mix_Data(dir_path=None)[source]

Bases: DatasetLoader

The Mix_Data dataset for social event detection.

Note

This dataset contains a mixture of social media data from various sources. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.CrisisLexT6(dir_path=None)[source]

Bases: DatasetLoader

The CrisisLexT6 dataset for social event detection.

Note

This dataset contains tweets related to 6 different crisis events. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset

class SocialED.dataset.dataloader.CrisisLexT7(dir_path=None)[source]

Bases: DatasetLoader

The CrisisLexT7 dataset for social event detection.

Note

This dataset contains tweets related to 7 different crisis events. The dataset provides text content and event labels for social event detection tasks.

load_data()[source]

Temporary implementation that returns empty dataset