# Dataset README

## Dataset for: Smartphone and web apps for pest and disease management in viticulture: a mapping of functionality, AI integration, and accessibility

**Associated publication:** Morelli, A., Masetti, A. Smartphone and Web Apps for Pest and Disease Management in Viticulture: A Mapping of Functionality, AI Integration, and Accessibility. *Smart Agricultural Technology*. https://doi.org/10.1016/j.atech.2026.102480 

**Authors / dataset creators:** Agata Morelli, Antonio Masetti — Department of Agricultural and Food Sciences, University of Bologna, Italy

**Contact:** agata.morelli2@unibo.it

**Funding:** This work was supported by the "Fondo per la Repubblica Digitale – Impresa Sociale" under the 'crescerAI' programme, supported by Google.org (Project code: 2023-CRE-00265).

**License:** This dataset is published under a Creative Commons Attribution 4.0 International license https://creativecommons.org/licenses/by/4.0/

---

## 1. Dataset description

This dataset is the application-level coding matrix underlying the mapping study described in the associated publication. It contains structured, binary and categorical variables coded for 49 smartphone and web-based applications supporting pest and disease management in viticulture, identified through a systematic search of commercial app stores (Google Play, Apple App Store) and Google Search conducted between March and May 2026 (search cutoff: 31 May 2026). Full details of the search, screening, and coding procedure are given in the Methods section of the associated manuscript.

Each row corresponds to one application. Each column corresponds to one coded variable.

## 2. File contents

| File | Description |
|---|---|
| `Dataset_repository_full_dataset.xlsx` | Two-sheet workbook: `Codebook` (variable dictionary and special-value legend) and `Apps` (49 data rows, 37 coded variables grouped into six thematic blocks, A–F, plus a source-link column). |

## 3. Data dictionary

Variables are grouped as they appear in the workbook (Section headers A–F correspond to the merged header row above the column names).

### A. General information

| Column | Type | Description |
|---|---|---|
| `App Name` | text | Application name as marketed by the developer. |
| `Primary Country` | categorical | Developer's primary country, determined from app store metadata, developer pages, company websites, or archived web history. Recorded as "Unknown" where this could not be confirmed (n = 1). |
| `Other Country` | text | A secondary developer country, recorded only where the application is associated with more than one country (e.g., co-development or dual headquarters). Blank for the majority of applications. |
| `Continent` | categorical | Continent associated with the developer's location. Used for the geographic distribution analysis (Section 3.1 of the manuscript). |
| `Year` | numeric | Year the application was first released, from public records. |
| `Year AI introduced` | numeric / "NA" | Year AI functionality was introduced, where applicable. Recorded as "NA" for applications with no AI functionality. |
| `Crops` | categorical | Whether the application is grapevine-exclusive or supports multiple crops. |
| `Platform` | categorical | Platform availability (iOS, Android, Web, or combinations). |
| `Category` | categorical | One of the four primary functional categories assigned via the hierarchical classification rule described in Section 2.3 of the manuscript. |

### B. Pest and disease coverage

| Column | Type | Description |
|---|---|---|
| `Pest/Disease Number` | numeric | Overall count of target pest/disease organisms disclosed by the developer. |
| `Organism species disclosed` | Yes/No | Whether the application disclosed a named list of pest/disease species (as distinct from an aggregate count only). |

### C. IPM functional components (binary, 1 = present / 0 = absent)

Each corresponds to one of the nine IPM components defined in Table 1 of the manuscript:

| Column | Corresponding IPM component |
|---|---|
| `ID/Diagnosis` | Disease or pest diagnosis and identification |
| `Monitoring` | Field monitoring and scouting |
| `Forecasting` | Disease or pest risk forecasting |
| `Management Options` | Management options and control measures |
| `Intervention Timing` | Intervention timing optimisation |
| `Precision control` | Precision and site-specific management |
| `Recordkeeping` | Recordkeeping and documentation |
| `Surveillance` | Surveillance or regulatory reporting |
| `Alerts` | Communication of alerts, warnings, or information |
| `IPM Score` | Sum of the nine binary components above (range 0–9). |

### D. Computational intelligence

| Column | Type | Description |
|---|---|---|
| `AI` | Yes/No | Whether the application is AI-enabled under the taxonomy (at least one of: image-based diagnosis/CV, ML-based risk forecasting, AI-driven advisory/conversational systems, AI-driven sensor-data interpretation). |
| `Evidence` | Confirmed/Inferred | Whether AI status was confirmed from technical detail in developer material, or inferred from functional description and terminology alone. |
| `Rule-based` | 0/1 | Presence of rule-based logic (recorded independently of AI status; not mutually exclusive with AI). |
| `Mecchanistic/Epidemiological Models` | 0/1 | Presence of mechanistic/epidemiological modelling. |
| `Weather Models` | 0/1 | Presence of weather-driven forecasting. |
| `AI CV` | 0/1 | AI functional type 1: image-based diagnosis / computer vision. |
| `AI Forecasting/ ML` | 0/1 | AI functional type 2: ML-based risk forecasting / predictive analytics. |
| `AI LLM` | 0/1 | AI functional type 3: advisory or conversational AI (e.g., LLM-based). |
| `AI Sensor Analytics` | 0/1 | AI functional type 4: AI-driven interpretation of sensor/IoT/trap data. |

Note: the four AI functional types are not mutually exclusive; an application may be coded 1 for more than one.

### E. Data and interoperability

| Column | Type | Description |
|---|---|---|
| `Open API` | 0/1 | Presence of a publicly documented, open/third-party API (even if entreprise API). |
| `Weather API` | 0/1 | Integration of external weather data sources. |
| `Export capability` | 0 / 1 / "Unknown" | Data export functionality in a standard format. "Unknown" indicates status could not be determined from public materials. |

### F. Pricing and adoption

| Column | Type | Description |
|---|---|---|
| `Price listed` | 0 / 1 / "No" | Whether subscription pricing is publicly disclosed. `"No"` is used for fully free applications (not applicable); `0`/`1` distinguish undisclosed vs. disclosed pricing among subscription/freemium applications. |
| `Free tier` | 0/1 | Whether any free-tier or freemium access exists, regardless of overall pricing classification. |
| `Pricing Model` | Free/Subscription | Binary pricing classification. Freemium and both subscription tiers (disclosed/undisclosed) are collapsed into "Subscription." |

### Sources

| Column | Type | Description |
|---|---|---|
| `Website` | URL | Official product/developer website; where no official website exists, the corresponding app store listing (Google Play or Apple App Store) is given instead.|


## 4. Data quality and limitations

> All information in this dataset was compiled from publicly available sources, including app store listings, developer websites, and associated promotional and technical materials, as described in the accompanying manuscript. While every entry reflects a good-faith reading of what developers themselves chose to disclose, some variation in how consistently and precisely that information was presented across platforms means occasional miscoding cannot be entirely ruled out. Given the scale of this mapping, applications were not individually downloaded or hands-on tested to verify every coded feature; this dataset should therefore be read as a systematic account of developer-published information, rather than an independently verified technical audit of each application.

## 5. Citation

If you use this dataset, please cite the associated publication:

> Morelli, A., Masetti, A. (2026) Smartphone and Web Apps for Pest and Disease Management in Viticulture: A Mapping of Functionality, AI Integration, and Accessibility. *Smart Agricultural Technology*. https://doi.org/10.1016/j.atech.2026.102480