LAMBDA / README.md
behavior-in-the-wild's picture
Update README.md (#5)
433f545 verified
|
raw
history blame
2.83 kB
metadata
dataset_info:
  features:
    - name: video_id
      dtype: int64
    - name: recall_score
      dtype: float64
    - name: youtube_id
      dtype: string
    - name: ad_details
      struct:
        - name: Audio
          dtype: string
        - name: Brand
          dtype: string
        - name: Duration
          dtype: string
        - name: Orientation
          dtype: string
        - name: Pace
          dtype: string
        - name: Scenes
          list:
            - name: Colors
              dtype: string
            - name: Description
              dtype: string
            - name: Emotions
              dtype: string
            - name: Number
              dtype: string
            - name: Photography Style
              dtype: string
            - name: Tags
              dtype: string
            - name: Text Shown
              dtype: string
            - name: Tone
              dtype: string
            - name: Visual Complexity
              dtype: string
        - name: Title
          dtype: string
  splits:
    - name: train
      num_bytes: 5490622.457169034
      num_examples: 1964
    - name: test
      num_bytes: 612243.5428309665
      num_examples: 219
  download_size: 2551503
  dataset_size: 6102866
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
license: mit
pretty_name: Long Term Memorability of Advertisements (LAMBDA)
task_categories:
  - text-classification
  - text-generation
  - question-answering
tags:
  - memorability
  - long-term-memorability
  - advertisement memorability

Dataset Description

Dataset Summary

LAMDBA is a long term ad memorability dataset, featuring data from 1749 participants and 2205 ads across 276 brands.

Dataset Structure

from datasets import load_dataset
ds = load_dataset("behavior-in-the-wild/LAMBDA")
ds

DatasetDict({
    train: Dataset({
        features: ['video_id', 'recall_score', 'youtube_id', 'ad_details'],
        num_rows: 1964
    })
    test: Dataset({
        features: ['video_id', 'recall_score', 'youtube_id', 'ad_details'],
        num_rows: 219
    })
})

Data Fields

  • video_id: identifier for the data sample
  • recall_score: memorability score for the video between 0 to 1
  • youtube_id: youtube id for the video
  • ad_details: scene by scene features for each video

Citation

@misc{s2024longtermadmemorabilityunderstanding, title={Long-Term Ad Memorability: Understanding and Generating Memorable Ads}, author={Harini S I au2 and Somesh Singh and Yaman K Singla and Aanisha Bhattacharyya and Veeky Baths and Changyou Chen and Rajiv Ratn Shah and Balaji Krishnamurthy}, year={2024}, eprint={2309.00378}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/2309.00378}}