BEET-M Quadrant-Mapped Verbs: A Fuzzy and binary Dataset of 9,000 English Verbs for Emotional and Behavioral Influence Modeling

- Citation Author(s):
- Submitted by:
- Wael Maged Badawy
- Last updated:
- DOI:
- 10.21227/43aj-8319
- Data Format:
Abstract
This dataset presents a curated collection of 9,000 English verbs annotated with normalized fuzzy values across four cognitive-behavioral quadrants of the BEET-M (Behavior Engagement Emotion Trigger Modes) model: Value & Credibility (NW), Relationship & Human Impact (NE), Process & Information (SE), and Time Urgency (SW). Each verb is assigned fuzzy scores summing to 1.0, along with a corresponding binary vector marking its dominant influence quadrant. This dataset supports applications in affective computing, persuasive communication, human-computer interaction, and natural language processing, particularly for emotionally adaptive systems and context-aware word scoring.
Instructions:
The dataset contains 9,000 rows with the following columns:
Column | Description |
---|---|
Verb | Base form of an English verb |
NW_fuzzy | Normalized fuzzy score for Value & Credibility quadrant |
NE_fuzzy | Normalized fuzzy score for Relationship & Human Impact quadrant |
SE_fuzzy | Normalized fuzzy score for Process & Information quadrant |
SW_fuzzy | Normalized fuzzy score for Time Urgency quadrant |
NW_binary | 1 if NW has the highest fuzzy score for the verb, else 0 |
NE_binary | 1 if NE is dominant |
SE_binary | 1 if SE is dominant |
SW_binary | 1 if SW is dominant |
All fuzzy scores are generated using a controlled random distribution normalized to sum to 1. Binary flags are computed using argmax()
logic.
Experimental Design, Materials and Methods
Data Source
We extracted verb forms from the open-source verbs-all.csv dataset, which includes base forms and conjugations. After deduplication and cleaning, the first column of each row (base form) was selected.
Fuzzy Value Generation
Each verb is assigned four fuzzy values generated using NumPy’s random distribution function and normalized across four dimensions:
pythonنسختحريرfuzzy = np.random.rand(4)
normalized = fuzzy / fuzzy.sum()
Binary Labeling
The quadrant with the highest fuzzy value was marked with a binary flag:
pythonنسختحريرdominant = np.argmax(normalized)
binary = [1 if i == dominant else 0 for i in range(4)]