Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,76 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
pretty_name: FRAPPE datset
|
| 6 |
+
size_categories:
|
| 7 |
+
- 10K<n<100K
|
| 8 |
---
|
| 9 |
+
|
| 10 |
+
**Dataset Description: Frappe Processed Dataset**
|
| 11 |
+
|
| 12 |
+
The Frappe dataset has been processed to refine the quality of user-item interactions by removing entries where either users or items had fewer than 5 interactions. This pruning resulted in a significant reduction in the dataset size:
|
| 13 |
+
|
| 14 |
+
- **Number of Users**: 651 (a reduction of 31.97% from the original dataset)
|
| 15 |
+
- **Number of Items**: 1127 (a reduction of 72.39%)
|
| 16 |
+
- **Total Number of Interactions**: 84,373 (a reduction of 12.30%)
|
| 17 |
+
|
| 18 |
+
**Columns Overview**:
|
| 19 |
+
The dataset includes 24 columns that capture a variety of features, structured as follows:
|
| 20 |
+
|
| 21 |
+
- **Basic Features**:
|
| 22 |
+
- `user`: User identifier
|
| 23 |
+
- `item`: Item identifier
|
| 24 |
+
- `rating`: Rating given by the user
|
| 25 |
+
- `cnt`: Count of interactions
|
| 26 |
+
|
| 27 |
+
- **Time of Day Features** (Encoded as one-hot):
|
| 28 |
+
- `daytime_afternoon`
|
| 29 |
+
- `daytime_evening`
|
| 30 |
+
- `daytime_morning`
|
| 31 |
+
- `daytime_night`
|
| 32 |
+
- `daytime_noon`
|
| 33 |
+
- `daytime_sunrise`
|
| 34 |
+
- `daytime_sunset`
|
| 35 |
+
|
| 36 |
+
- **Weather Conditions** (Encoded as one-hot):
|
| 37 |
+
- `weather_cloudy`
|
| 38 |
+
- `weather_drizzle`
|
| 39 |
+
- `weather_foggy`
|
| 40 |
+
- `weather_rainy`
|
| 41 |
+
- `weather_snowy`
|
| 42 |
+
- `weather_stormy`
|
| 43 |
+
- `weather_sunny`
|
| 44 |
+
- `weather_unknown`
|
| 45 |
+
|
| 46 |
+
- **Weekend Indicator**:
|
| 47 |
+
- `isweekend_weekend`
|
| 48 |
+
- `isweekend_workday`
|
| 49 |
+
|
| 50 |
+
- **Location of Use (Home or Work)**:
|
| 51 |
+
- `homework_home`
|
| 52 |
+
- `homework_unknown`
|
| 53 |
+
- `homework_work`
|
| 54 |
+
|
| 55 |
+
This dataset is designed for use in building Context-Aware Recommender Systems (CARS), offering rich features that capture temporal dynamics, environmental conditions, and user contexts, which are critical for personalized recommendations.
|
| 56 |
+
|
| 57 |
+
**Any scientific publications that use this data set should cite the following paper as the reference:**
|
| 58 |
+
@Article{frappe15,
|
| 59 |
+
title={Frappe: Understanding the Usage and Perception of Mobile App Recommendations In-The-Wild},
|
| 60 |
+
author = {Linas Baltrunas, Karen Church, Alexandros Karatzoglou, Nuria Oliver},
|
| 61 |
+
date={2015},
|
| 62 |
+
urldate={2015-05-12},
|
| 63 |
+
eprinttype={arxiv},
|
| 64 |
+
eprint={arXiv:1505.03014}
|
| 65 |
+
}
|
| 66 |
+
|
| 67 |
+
Nobody guarantees the correctness of the data, its suitability for any particular purpose,
|
| 68 |
+
or the validity of results based on the use of the data set. The data set may be used for any
|
| 69 |
+
research purposes under the following conditions:
|
| 70 |
+
* The user must acknowledge the use of the data set in publications resulting from the use of the data set.
|
| 71 |
+
* The user may not redistribute the data without separate permission.
|
| 72 |
+
* The user may not try to deanonymise the data.
|
| 73 |
+
* The user may not use this information for any commercial or revenue-bearing purposes without first obtaining permission from us.
|
| 74 |
+
|
| 75 |
+
In no event anyone involved in frappe project be liable to you for any damages arising out of the use or inability to use the
|
| 76 |
+
associated scripts (including but not limited to loss of data or data being rendered inaccurate).
|