abadesalex/Frappe-mobile-app-usage

---
license: apache-2.0
language:
- en
pretty_name: FRAPPE datset
size_categories:
- 10K<n<100K
---

**Dataset Description: Frappe Processed Dataset**

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:

- **Number of Users**: 651 (a reduction of 31.97% from the original dataset)
- **Number of Items**: 1127 (a reduction of 72.39%)
- **Total Number of Interactions**: 84,373 (a reduction of 12.30%)

**Columns Overview**:
The dataset includes 24 columns that capture a variety of features, structured as follows:

- **Basic Features**:
- `user`: User identifier
- `item`: Item identifier
- `rating`: Rating given by the user
- `cnt`: Count of interactions

- **Time of Day Features** (Encoded as one-hot):
- `daytime_afternoon`
- `daytime_evening`
- `daytime_morning`
- `daytime_night`
- `daytime_noon`
- `daytime_sunrise`
- `daytime_sunset`

- **Weather Conditions** (Encoded as one-hot):
- `weather_cloudy`
- `weather_drizzle`
- `weather_foggy`
- `weather_rainy`
- `weather_snowy`
- `weather_stormy`
- `weather_sunny`
- `weather_unknown`

- **Weekend Indicator**:
- `isweekend_weekend`
- `isweekend_workday`

- **Location of Use (Home or Work)**:
- `homework_home`
- `homework_unknown`
- `homework_work`

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.

**Any scientific publications that use this data set should cite the following paper as the reference:**
@Article{frappe15,
title={Frappe: Understanding the Usage and Perception of Mobile App Recommendations In-The-Wild},
author = {Linas Baltrunas, Karen Church, Alexandros Karatzoglou, Nuria Oliver},
date={2015},
urldate={2015-05-12},
eprinttype={arxiv},
eprint={arXiv:1505.03014}
}

Nobody guarantees the correctness of the data, its suitability for any particular purpose,
or the validity of results based on the use of the data set. The data set may be used for any
research purposes under the following conditions:
* The user must acknowledge the use of the data set in publications resulting from the use of the data set.
* The user may not redistribute the data without separate permission.
* The user may not try to deanonymise the data.
* The user may not use this information for any commercial or revenue-bearing purposes without first obtaining permission from us.

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
associated scripts (including but not limited to loss of data or data being rendered inaccurate).

license: apache-2.0
language:
- 英语
pretty_name: FRAPPE 数据集
size_categories:
- 10000 < 数据规模 < 100000

**数据集描述：经过预处理的Frappe数据集**

本次预处理通过剔除交互次数不足5次的用户与物品条目，以优化用户-物品交互数据的质量。经此修剪操作后，数据集规模出现显著缩减：
- **用户总数**：651位（较原始数据集减少31.97%）
- **物品总数**：1127个（较原始数据集减少72.39%）
- **总交互数**：84373条（较原始数据集减少12.30%）

**字段概览**
本数据集共包含24个字段，涵盖多维度特征，结构如下：

### 基础特征
- `user`：用户标识符
- `item`：物品标识符
- `rating`：用户给出的评分
- `cnt`：交互次数

### 时段特征（采用独热编码（one-hot））
- `daytime_afternoon`：午后时段
- `daytime_evening`：傍晚时段
- `daytime_morning`：清晨时段
- `daytime_night`：夜间时段
- `daytime_noon`：正午时段
- `daytime_sunrise`：日出时段
- `daytime_sunset`：日落时段

### 天气状况特征（采用独热编码（one-hot））
- `weather_cloudy`：多云天气
- `weather_drizzle`：毛毛雨天气
- `weather_foggy`：有雾天气
- `weather_rainy`：降雨天气
- `weather_snowy`：降雪天气
- `weather_stormy`：风暴天气
- `weather_sunny`：晴朗天气
- `weather_unknown`：未知天气

### 周末标识特征
- `isweekend_weekend`：周末场景
- `isweekend_workday`：工作日场景

### 使用位置特征（家庭或工作场所）
- `homework_home`：家庭场景
- `homework_unknown`：未知场景
- `homework_work`：工作场景

本数据集专为构建**上下文感知推荐系统（Context-Aware Recommender Systems, CARS）**而设计，提供了涵盖时间动态、环境条件与用户上下文的丰富特征，这些特征对于个性化推荐任务至关重要。

**引用规范**
使用本数据集的学术出版物需引用以下文献：
bibtex
@Article{frappe15,
title={Frappe: Understanding the Usage and Perception of Mobile App Recommendations In-The-Wild},
author = {Linas Baltrunas, Karen Church, Alexandros Karatzoglou, Nuria Oliver},
date={2015},
urldate={2015-05-12},
eprinttype={arxiv},
eprint={arXiv:1505.03014}
}

本数据集不保证数据的正确性、适用于特定用途的合理性，以及基于该数据集所得结果的有效性。该数据集可在以下条件下用于各类研究用途：
* 使用者需在使用该数据集产生的学术出版物中注明数据集的使用情况；
* 未经单独许可，使用者不得重新分发该数据集；
* 使用者不得尝试对数据集进行去匿名化处理；
* 未经我方许可，使用者不得将本数据集用于任何商业或营利性用途。

在任何情况下，Frappe项目相关人员均不对因使用或无法使用相关脚本而产生的任何损害承担责任（包括但不限于数据丢失或数据不准确）。

原始信息汇总

数据集概述

数据集名称: FRAPPE Processed Dataset

数据集描述: 该数据集经过处理，通过移除用户或物品少于5次交互的条目，提高了用户-物品交互的质量。处理后的数据集大小如下：

• 用户数量: 651（相比原始数据集减少了31.97%）
• 物品数量: 1127（相比原始数据集减少了72.39%）
• 总交互次数: 84,373（相比原始数据集减少了12.30%）

数据集结构: 包含24个列，涵盖了多种特征，具体包括：

• 基本特征:

  • user: 用户标识
  • item: 物品标识
  • rating: 用户给出的评分
  • cnt: 交互次数

• 时间特征（独热编码）:

  • daytime_afternoon
  • daytime_evening
  • daytime_morning
  • daytime_night
  • daytime_noon
  • daytime_sunrise
  • daytime_sunset

• 天气条件（独热编码）:

  • weather_cloudy
  • weather_drizzle
  • weather_foggy
  • weather_rainy
  • weather_snowy
  • weather_stormy
  • weather_sunny
  • weather_unknown

• 周末指示:

  • isweekend_weekend
  • isweekend_workday

• 使用地点（家或工作）:

  • homework_home
  • homework_unknown
  • homework_work

数据集用途: 该数据集专为构建上下文感知推荐系统（CARS）设计，提供了丰富的特征，包括时间动态、环境条件和用户上下文，这些对于个性化推荐至关重要。

引用信息: 使用此数据集的科学出版物应引用以下论文：

@Article{frappe15, title={Frappe: Understanding the Usage and Perception of Mobile App Recommendations In-The-Wild}, author = {Linas Baltrunas, Karen Church, Alexandros Karatzoglou, Nuria Oliver}, date={2015}, urldate={2015-05-12}, eprinttype={arxiv}, eprint={arXiv:1505.03014} }