A dataset optimized for detecting activities such as falls, walking, and jogging. It is primarily used in creating apps that use smartphone sensors to detect falls, particularly in elderly individuals.
Although not exclusively designed for HAR, the NHANES dataset is a rich source of health and nutritional data, which could potentially be utilized to garner insights into human activities and health conditions.
This repository houses data concerning human activities collected from smartphone accelerometer sensors. It serves as a valuable resource for developing machine learning models capable of recognizing various activities.
4.**[UCI Human Activity Recognition Using Smartphones Dataset](https://archive.ics.uci.edu/dataset/240/human+activity+recognition+using+smartphones)**
This dataset comprises data from smartphone accelerometers and gyroscopes, capturing activities such as walking, sitting, and standing performed by 30 subjects. It is a popular choice for HAR research projects.
5.**[ISDM (Wireless Sensor Data Mining)](https://github.com/topics/wireless-sensor-data-mining)**
Although not a database per se, this GitHub topic connects you to various resources and datasets pertaining to wireless sensor data mining, an essential aspect in HAR research.
6.**[HHAR (Heterogeneity Human Activity Recognition)](https://github.com/Limmen/Distributed_ML)**
HHAR stands out with its data collected from a range of devices, portraying various human activities. It is particularly beneficial for constructing models adaptable to different data sources.
Featuring data from wearable sensors monitoring individuals performing diverse physical activities, PAMAP2 is a vital tool for developing predictive HAR models.
8.**[Daphnet Freezing of Gait](https://archive.ics.uci.edu/dataset/245/daphnet+freezing+of+gait)**
Focused on Parkinson's patients' gait freezing, this dataset, comprising data from wearable sensors, plays a crucial role in HAR healthcare applications.
Developed to recognize various physical activities through smartphone sensors, Actitracker houses data on activities such as walking and jogging.
10.**[Daily and Sports Activities](https://archive.ics.uci.edu/dataset/256/daily+and+sports+activities)**
This dataset contains data on a range of daily and sports activities recorded through wearable sensors, making it a rich resource for HAR research, especially in distinguishing between different physical activities.
11.**[Smartphone Dataset for Human Activity Recognition (HAR) in Ambient Assisted Living (AAL)](https://archive.ics.uci.edu/dataset/364/smartphone+dataset+for+human+activity+recognition+har+in+ambient+assisted+living+aal)**
This dataset focuses on aiding the elderly or disabled, using smartphone sensors to identify their activities, hence fostering safer and more comfortable living environments.
This dataset is notable for its emphasis on context recognition, using sensor data from various sources to identify complex activities and gestures, thereby advancing research in ambient intelligence.
13.**[CASAS](https://casas.wsu.edu/datasets/)**
CASAS, a collection of datasets centered on smart home environments, facilitates the creation of algorithms capable of recognizing home-based activities through sensor data.
This dataset distinguishes itself with its inclusion of depth maps alongside skeletal data for activity recognition, aiding in the development of models capable of identifying activities from 3D data.
REALDISP incorporates data on various activities captured through wearable sensors, with a focus on realistic data disposition, which is vital for creating robust HAR models.
