Datasets:

mohitg2
/

SenAffLex-Provo

# Processed Provo Corpus — Licensing, Attribution, and Citation

## SenAffLex-Provo Research Repository

**Repository:** SenAffLex-Provo  
**Research Resource:** Intermediate Provo Processing Outputs  
**Computational Processing and Implementation:** Mohit Gupta  
**Repository Creator and Maintainer:** Mohit Gupta  
**Institution:** University of Illinois Urbana-Champaign  
**Year:** 2026  
**License:** CC BY 4.0 — eligible original contributions in Files 3 and 4 only  
**Documentation Version:** 1.3  
**Last Updated:** October 2026  

**Repository URL:** https://huggingface.co/datasets/mohitg2/SenAffLex-Provo  
**Repository DOI:** {Pending Repository DOI assignment}

---

## IMPORTANT — LICENSE SCOPE

**This document licenses only eligible original contributions by Mohit Gupta in the following two computationally generated vocabulary-coverage files:**

- `3_provo_missing_vocabulary_pre_lemmatization.csv`
- `4_provo_missing_vocabulary_post_lemmatization.csv`

**Files 1 and 2 were also processed by Mohit Gupta, but they are expressly excluded from the CC BY 4.0 grant in this document:**

- `1_Provo_Norming_Item_Level_Cloze.csv`
- `2_processed_Provo_Corpus.csv`

Files 1 and 2 contain processed information originating from the Provo Corpus. Their exclusion does **not** deny or diminish Mohit Gupta's computational processing contributions; it reflects the need to verify applicable source-data redistribution permissions. This document grants no new permission to redistribute those two files.

**The license for Files 3 and 4 covers only contributions that Mohit Gupta has the rights to license.** It does not relicense underlying Provo Corpus words, observations, predictability norms, or other third-party materials that may appear in those reports.

The fact that a report was computationally generated does not automatically establish copyright in all of its contents.

---

## 1. Overview

This directory documents four intermediate outputs produced during the preparation of Provo-derived research data for the SenAffLex-Provo research program.

Mohit Gupta developed and implemented the computational procedures used to process or generate all four files.

The workflow includes:

- Data cleaning and standardization.
- Restructuring of predictability information.
- Token-level data preparation.
- Vocabulary extraction and matching.
- Identification of unmatched lexical items.
- Lemmatization-related comparisons.
- Dataset validation.

These intermediate outputs support the subsequent construction of:

1. Master Provo.
2. Lexical Provo.
3. Affective Provo.

Integration of GLENS sentiment norms into a future SenAffLex-Provo dataset is planned but is not represented as completed in this document.

---

## 2. Dataset Inventory and Licensing Status

| File | Processing Contribution | Status Under This Document |
|---|---|---|
| `1_Provo_Norming_Item_Level_Cloze.csv` | Item-level cloze data preparation, restructuring, and validation by Mohit Gupta. | **Excluded from this license.** Source-data redistribution permissions remain to be verified. |
| `2_processed_Provo_Corpus.csv` | Token-level Provo data cleaning, standardization, restructuring, and validation by Mohit Gupta. | **Excluded from this license.** Source-data redistribution permissions remain to be verified. |
| `3_provo_missing_vocabulary_pre_lemmatization.csv` | Computational vocabulary matching and identification of unmatched items before lemmatization. | **CC BY 4.0 for eligible original contributions only**, subject to third-party rights. |
| `4_provo_missing_vocabulary_post_lemmatization.csv` | Lemmatization-related vocabulary matching and coverage analysis. | **CC BY 4.0 for eligible original contributions only**, subject to third-party rights. |

**Availability is separate from licensing.**

This document does not claim that any file is currently public, private, removed from Git history, or authorized for redistribution.

Repository availability should be checked independently.

---

## 3. Original Computational Contributions

The original computational implementation was developed by **Mohit Gupta** and includes, where applicable:

1. Preparing and standardizing Provo-derived source files.
2. Restructuring item-level cloze predictability data.
3. Preparing token-level research observations.
4. Extracting and matching lexical items.
5. Identifying unmatched vocabulary before lemmatization.
6. Implementing lemmatization-related vocabulary comparisons.
7. Generating coverage reports and intermediate outputs.
8. Validating the resulting files and documenting the workflow.

These statements concern the computational implementation, not ownership of the original Provo Corpus or authorship of related scientific publications.

The corresponding original preprocessing code is licensed separately under MIT where designated.

---

## 4. CC BY 4.0 License for Files 3 and 4

Eligible original contributions by Mohit Gupta contained in **Files 3 and 4 only** are offered under the **Creative Commons Attribution 4.0 International License (CC BY 4.0)**, to the extent that those contributions are protected by applicable rights and Mohit Gupta has authority to license them.

**Official License:**  
https://creativecommons.org/licenses/by/4.0/

For covered material, CC BY 4.0 permits:

- Sharing and redistribution.
- Adaptation and modification.
- Incorporation into other research resources.
- Commercial and noncommercial reuse.

Users must comply with the applicable CC BY 4.0 conditions, including:

1. Providing appropriate attribution.
2. Providing a link to the license.
3. Indicating modifications where applicable.
4. Not implying endorsement by the original creator.

**This license does not apply to Files 1 and 2.**

It also does not override the rights or distribution conditions of the original Provo Corpus or other third-party resources, including material embedded in Files 3 and 4.

No additional mandatory academic-publication citation condition is imposed beyond the standard CC BY 4.0 terms.

The scholarly citation request below is separate from legal license obligations.

---

## 5. Original Provo Corpus Attribution

The intermediate outputs originate from or refer to the Provo Corpus developed by Luke and Christianson.

### Original Publication

Luke, S. G., & Christianson, K. (2018). The Provo Corpus: A large eye-tracking corpus with predictability norms. *Behavior Research Methods, 50*, 826–833.

https://doi.org/10.3758/s13428-017-0908-4

Researchers using Provo-derived material should cite the original publication.

Original observations, predictability norms, and other source materials remain subject to applicable source rights and permissions.

**Redistribution Permission Status for Files 1 and 2:** Verification pending.

Their processing by Mohit Gupta does not itself establish permission to redistribute the underlying source data.

---

## 6. Scholarly Citation of SenAffLex-Provo

Researchers who substantially use the original vocabulary-coverage reports, preprocessing methods, or related computational resources are **strongly requested to cite the repository**.

### Recommended Repository Citation

Gupta, M. (2026). *SenAffLex-Provo: Integrated sentiment, affective, and lexical research datasets and computational resources* [Data set and software]. Hugging Face. {Pending Repository DOI assignment}

**Repository URL:**  
https://huggingface.co/datasets/mohitg2/SenAffLex-Provo

The placeholder in curly braces will be replaced by the full DOI URL after assignment.

The repository's current released resources are lexical and affective; sentiment integration is planned.

### Associated Research Publication

Researchers using findings or analyses from the associated article should also cite:

Gupta, M., & Christianson, K. (in press). Beyond word length, frequency, and predictability: Affective properties predict eye movement in reading. *Reading Research Quarterly*.

**Publication Status:** Accepted October 20, 2026.  
**Article DOI:** Pending assignment.

The repository citation, article citation, and original Provo Corpus citation recognize distinct contributions and should be used as relevant.

Formal scholarly citation is requested but is not an additional legal condition of CC BY 4.0.

---

## 7. Suggested Attribution

For eligible original contributions in Files 3 and 4, researchers may use the following acknowledgment, adapted as appropriate:

> Vocabulary-coverage reports were generated using preprocessing and lemmatization-related computational procedures developed by Mohit Gupta for the SenAffLex-Provo research program at the University of Illinois Urbana-Champaign. Source vocabulary was derived from the Provo Corpus (Luke & Christianson, 2018).

This is suggested wording, not a required formula.

Users must still satisfy any applicable license-attribution obligations.

---

## 8. Redistribution and Third-Party Rights

### 8.1 Files 1 and 2

The following files are excluded from this licensing grant:

- `1_Provo_Norming_Item_Level_Cloze.csv`
- `2_processed_Provo_Corpus.csv`

Although Mohit Gupta developed and implemented their computational processing procedures, this document does not grant permission to redistribute their complete contents.

The applicable original source-data permissions must be established separately.

### 8.2 Files 3 and 4

The following files contain eligible original computational contributions offered under CC BY 4.0:

- `3_provo_missing_vocabulary_pre_lemmatization.csv`
- `4_provo_missing_vocabulary_post_lemmatization.csv`

Any third-party material contained within these files remains subject to its applicable rights and terms.

### 8.3 Original Computational Software

Original Python and R code is governed by its separately designated software license, generally MIT for original SenAffLex-Provo code.

### 8.4 General Redistribution Conditions

Selecting, cleaning, lemmatizing, or reorganizing factual data does not automatically establish exclusive rights in the resulting values.

Users are responsible for checking applicable third-party conditions before redistribution.

---

## 9. Reproducibility and Access

The computational pipeline documents the preparation of all four intermediate files.

Researchers may reproduce the processing workflow using appropriately obtained source data and the published preprocessing scripts, subject to applicable access and use conditions.

Where direct redistribution of a source-derived file is not authorized, the repository may provide:

- Original source-access instructions.
- Preprocessing scripts.
- Vocabulary matching procedures.
- Lemmatization workflows.
- Dataset reconstruction instructions.
- Validation procedures.

Such instructions do not themselves grant rights to the original data.

---

## 10. Additional Source Citations

The broader SenAffLex-Provo research program also uses:

- SUBTLEX-US frequency information.
- Warriner et al. (2013) affective norms.

Their citations and source-rights information are documented in:

[Original Data Sources and Citations](../01-Raw/RawDataCitations.md)

Their inclusion in broader project documentation does not mean that every intermediate file contains those resources.

---

## 11. Updates and Versioning

This document may be updated to record:

- Verified source-data permissions.
- Dataset revisions.
- Associated research publications.
- Repository DOI assignments.
- Journal article DOI assignments.
- Updated citation information.
- Additional computational contributions.

Any later licensing changes will respect existing valid grants and third-party rights.

Once assigned, replace:

`{Pending Repository DOI assignment}`

with the repository DOI URL.

Researchers should identify the specific repository release or Git commit used for reproducibility.

---

## 12. Repository Information

**Repository:** SenAffLex-Provo  
**Research Resource:** Intermediate Provo Processing Outputs  
**Original Computational Developer:** Mohit Gupta  
**Computational Processing and Implementation:** Mohit Gupta  
**Institution:** University of Illinois Urbana-Champaign  

**Repository URL:**  
https://huggingface.co/datasets/mohitg2/SenAffLex-Provo

**Repository DOI:** {Pending Repository DOI assignment}

**CC BY 4.0 Scope:** Eligible original contributions in Files 3 and 4 only.

**Files 1 and 2:** Processed by Mohit Gupta; excluded from this license.

**Original Provo Corpus:** Original source rights and permissions apply.

**Documentation Version:** 1.3  
**Last Updated:** October 2026

---

## 13. Important Legal Notice

**Only eligible original contributions in `3_provo_missing_vocabulary_pre_lemmatization.csv` and `4_provo_missing_vocabulary_post_lemmatization.csv` are offered under CC BY 4.0 by this document.**

**`1_Provo_Norming_Item_Level_Cloze.csv` and `2_processed_Provo_Corpus.csv` were processed by Mohit Gupta but are excluded from this licensing grant.**

This document neither claims ownership of original Provo Corpus material nor grants permission to redistribute third-party data.

Citation and attribution do not substitute for any legally required redistribution permission.

	# Processed Provo Corpus — Licensing, Attribution, and Citation

	## SenAffLex-Provo Research Repository

	Repository: SenAffLex-Provo
	Research Resource: Intermediate Provo Processing Outputs
	Computational Processing and Implementation: Mohit Gupta
	Repository Creator and Maintainer: Mohit Gupta
	Institution: University of Illinois Urbana-Champaign
	Year: 2026
	License: CC BY 4.0 — eligible original contributions in Files 3 and 4 only
	Documentation Version: 1.3
	Last Updated: October 2026

	Repository URL: https://huggingface.co/datasets/mohitg2/SenAffLex-Provo
	Repository DOI: {Pending Repository DOI assignment}

	---

	## IMPORTANT — LICENSE SCOPE

	This document licenses only eligible original contributions by Mohit Gupta in the following two computationally generated vocabulary-coverage files:

	- `3_provo_missing_vocabulary_pre_lemmatization.csv`
	- `4_provo_missing_vocabulary_post_lemmatization.csv`

	Files 1 and 2 were also processed by Mohit Gupta, but they are expressly excluded from the CC BY 4.0 grant in this document:

	- `1_Provo_Norming_Item_Level_Cloze.csv`
	- `2_processed_Provo_Corpus.csv`

	Files 1 and 2 contain processed information originating from the Provo Corpus. Their exclusion does not deny or diminish Mohit Gupta's computational processing contributions; it reflects the need to verify applicable source-data redistribution permissions. This document grants no new permission to redistribute those two files.

	The license for Files 3 and 4 covers only contributions that Mohit Gupta has the rights to license. It does not relicense underlying Provo Corpus words, observations, predictability norms, or other third-party materials that may appear in those reports.

	The fact that a report was computationally generated does not automatically establish copyright in all of its contents.

	---

	## 1. Overview

	This directory documents four intermediate outputs produced during the preparation of Provo-derived research data for the SenAffLex-Provo research program.

	Mohit Gupta developed and implemented the computational procedures used to process or generate all four files.

	The workflow includes:

	- Data cleaning and standardization.
	- Restructuring of predictability information.
	- Token-level data preparation.
	- Vocabulary extraction and matching.
	- Identification of unmatched lexical items.
	- Lemmatization-related comparisons.
	- Dataset validation.

	These intermediate outputs support the subsequent construction of:

	1. Master Provo.
	2. Lexical Provo.
	3. Affective Provo.

	Integration of GLENS sentiment norms into a future SenAffLex-Provo dataset is planned but is not represented as completed in this document.

	---

	## 2. Dataset Inventory and Licensing Status

	\| File \| Processing Contribution \| Status Under This Document \|
	\|---\|---\|---\|
	\| `1_Provo_Norming_Item_Level_Cloze.csv` \| Item-level cloze data preparation, restructuring, and validation by Mohit Gupta. \| Excluded from this license. Source-data redistribution permissions remain to be verified. \|
	\| `2_processed_Provo_Corpus.csv` \| Token-level Provo data cleaning, standardization, restructuring, and validation by Mohit Gupta. \| Excluded from this license. Source-data redistribution permissions remain to be verified. \|
	\| `3_provo_missing_vocabulary_pre_lemmatization.csv` \| Computational vocabulary matching and identification of unmatched items before lemmatization. \| CC BY 4.0 for eligible original contributions only, subject to third-party rights. \|
	\| `4_provo_missing_vocabulary_post_lemmatization.csv` \| Lemmatization-related vocabulary matching and coverage analysis. \| CC BY 4.0 for eligible original contributions only, subject to third-party rights. \|

	Availability is separate from licensing.

	This document does not claim that any file is currently public, private, removed from Git history, or authorized for redistribution.

	Repository availability should be checked independently.

	---

	## 3. Original Computational Contributions

	The original computational implementation was developed by Mohit Gupta and includes, where applicable:

	1. Preparing and standardizing Provo-derived source files.
	2. Restructuring item-level cloze predictability data.
	3. Preparing token-level research observations.
	4. Extracting and matching lexical items.
	5. Identifying unmatched vocabulary before lemmatization.
	6. Implementing lemmatization-related vocabulary comparisons.
	7. Generating coverage reports and intermediate outputs.
	8. Validating the resulting files and documenting the workflow.

	These statements concern the computational implementation, not ownership of the original Provo Corpus or authorship of related scientific publications.

	The corresponding original preprocessing code is licensed separately under MIT where designated.

	---

	## 4. CC BY 4.0 License for Files 3 and 4

	Eligible original contributions by Mohit Gupta contained in Files 3 and 4 only are offered under the Creative Commons Attribution 4.0 International License (CC BY 4.0), to the extent that those contributions are protected by applicable rights and Mohit Gupta has authority to license them.

	Official License:
	https://creativecommons.org/licenses/by/4.0/

	For covered material, CC BY 4.0 permits:

	- Sharing and redistribution.
	- Adaptation and modification.
	- Incorporation into other research resources.
	- Commercial and noncommercial reuse.

	Users must comply with the applicable CC BY 4.0 conditions, including:

	1. Providing appropriate attribution.
	2. Providing a link to the license.
	3. Indicating modifications where applicable.
	4. Not implying endorsement by the original creator.

	This license does not apply to Files 1 and 2.

	It also does not override the rights or distribution conditions of the original Provo Corpus or other third-party resources, including material embedded in Files 3 and 4.

	No additional mandatory academic-publication citation condition is imposed beyond the standard CC BY 4.0 terms.

	The scholarly citation request below is separate from legal license obligations.

	---

	## 5. Original Provo Corpus Attribution

	The intermediate outputs originate from or refer to the Provo Corpus developed by Luke and Christianson.

	### Original Publication

	Luke, S. G., & Christianson, K. (2018). The Provo Corpus: A large eye-tracking corpus with predictability norms. Behavior Research Methods, 50, 826–833.

	https://doi.org/10.3758/s13428-017-0908-4

	Researchers using Provo-derived material should cite the original publication.

	Original observations, predictability norms, and other source materials remain subject to applicable source rights and permissions.

	Redistribution Permission Status for Files 1 and 2: Verification pending.

	Their processing by Mohit Gupta does not itself establish permission to redistribute the underlying source data.

	---

	## 6. Scholarly Citation of SenAffLex-Provo

	Researchers who substantially use the original vocabulary-coverage reports, preprocessing methods, or related computational resources are strongly requested to cite the repository.

	### Recommended Repository Citation

	Gupta, M. (2026). SenAffLex-Provo: Integrated sentiment, affective, and lexical research datasets and computational resources [Data set and software]. Hugging Face. {Pending Repository DOI assignment}

	Repository URL:
	https://huggingface.co/datasets/mohitg2/SenAffLex-Provo

	The placeholder in curly braces will be replaced by the full DOI URL after assignment.

	The repository's current released resources are lexical and affective; sentiment integration is planned.

	### Associated Research Publication

	Researchers using findings or analyses from the associated article should also cite:

	Gupta, M., & Christianson, K. (in press). Beyond word length, frequency, and predictability: Affective properties predict eye movement in reading. Reading Research Quarterly.

	Publication Status: Accepted October 20, 2026.
	Article DOI: Pending assignment.

	The repository citation, article citation, and original Provo Corpus citation recognize distinct contributions and should be used as relevant.

	Formal scholarly citation is requested but is not an additional legal condition of CC BY 4.0.

	---

	## 7. Suggested Attribution

	For eligible original contributions in Files 3 and 4, researchers may use the following acknowledgment, adapted as appropriate:

	> Vocabulary-coverage reports were generated using preprocessing and lemmatization-related computational procedures developed by Mohit Gupta for the SenAffLex-Provo research program at the University of Illinois Urbana-Champaign. Source vocabulary was derived from the Provo Corpus (Luke & Christianson, 2018).

	This is suggested wording, not a required formula.

	Users must still satisfy any applicable license-attribution obligations.

	---

	## 8. Redistribution and Third-Party Rights

	### 8.1 Files 1 and 2

	The following files are excluded from this licensing grant:

	- `1_Provo_Norming_Item_Level_Cloze.csv`
	- `2_processed_Provo_Corpus.csv`

	Although Mohit Gupta developed and implemented their computational processing procedures, this document does not grant permission to redistribute their complete contents.

	The applicable original source-data permissions must be established separately.

	### 8.2 Files 3 and 4

	The following files contain eligible original computational contributions offered under CC BY 4.0:

	- `3_provo_missing_vocabulary_pre_lemmatization.csv`
	- `4_provo_missing_vocabulary_post_lemmatization.csv`

	Any third-party material contained within these files remains subject to its applicable rights and terms.

	### 8.3 Original Computational Software

	Original Python and R code is governed by its separately designated software license, generally MIT for original SenAffLex-Provo code.

	### 8.4 General Redistribution Conditions

	Selecting, cleaning, lemmatizing, or reorganizing factual data does not automatically establish exclusive rights in the resulting values.

	Users are responsible for checking applicable third-party conditions before redistribution.

	---

	## 9. Reproducibility and Access

	The computational pipeline documents the preparation of all four intermediate files.

	Researchers may reproduce the processing workflow using appropriately obtained source data and the published preprocessing scripts, subject to applicable access and use conditions.

	Where direct redistribution of a source-derived file is not authorized, the repository may provide:

	- Original source-access instructions.
	- Preprocessing scripts.
	- Vocabulary matching procedures.
	- Lemmatization workflows.
	- Dataset reconstruction instructions.
	- Validation procedures.

	Such instructions do not themselves grant rights to the original data.

	---

	## 10. Additional Source Citations

	The broader SenAffLex-Provo research program also uses:

	- SUBTLEX-US frequency information.
	- Warriner et al. (2013) affective norms.

	Their citations and source-rights information are documented in:

	[Original Data Sources and Citations](../01-Raw/RawDataCitations.md)

	Their inclusion in broader project documentation does not mean that every intermediate file contains those resources.

	---

	## 11. Updates and Versioning

	This document may be updated to record:

	- Verified source-data permissions.
	- Dataset revisions.
	- Associated research publications.
	- Repository DOI assignments.
	- Journal article DOI assignments.
	- Updated citation information.
	- Additional computational contributions.

	Any later licensing changes will respect existing valid grants and third-party rights.

	Once assigned, replace:

	`{Pending Repository DOI assignment}`

	with the repository DOI URL.

	Researchers should identify the specific repository release or Git commit used for reproducibility.

	---

	## 12. Repository Information

	Repository: SenAffLex-Provo
	Research Resource: Intermediate Provo Processing Outputs
	Original Computational Developer: Mohit Gupta
	Computational Processing and Implementation: Mohit Gupta
	Institution: University of Illinois Urbana-Champaign

	Repository URL:
	https://huggingface.co/datasets/mohitg2/SenAffLex-Provo

	Repository DOI: {Pending Repository DOI assignment}

	CC BY 4.0 Scope: Eligible original contributions in Files 3 and 4 only.

	Files 1 and 2: Processed by Mohit Gupta; excluded from this license.

	Original Provo Corpus: Original source rights and permissions apply.

	Documentation Version: 1.3
	Last Updated: October 2026

	---

	## 13. Important Legal Notice

	Only eligible original contributions in `3_provo_missing_vocabulary_pre_lemmatization.csv` and `4_provo_missing_vocabulary_post_lemmatization.csv` are offered under CC BY 4.0 by this document.

	`1_Provo_Norming_Item_Level_Cloze.csv` and `2_processed_Provo_Corpus.csv` were processed by Mohit Gupta but are excluded from this licensing grant.

	This document neither claims ownership of original Provo Corpus material nor grants permission to redistribute third-party data.

	Citation and attribution do not substitute for any legally required redistribution permission.