Machine Learning

The Deepfake face detection task involves a facial image of unknown authenticity for testing. While most deepfake detection methods take only the image as input, our literature demonstrates that conditioning the deepfake detector on identity—i.e., knowing whose deepfake face the picture might be—can enhance detection performance. Existing deepfake detection datasets, such as FaceForensics++ and DFDC, do not include identity information for authentic and deepfake faces.

Categories:
1269 Views

The dataset utilized in this research originates from two primary sources: the Central Bureau of Statistics of Indonesia, which provides data on Harvested Area and Production, and the Meteorology, Climatology, and Geophysics Agency of Indonesia, responsible for data on Rainfall, Humidity, and Temperature. This dataset encompasses six years of observations, collected annually from 2018 to 2023. It is important to note that the data for December 2023 are predictive estimates from these agencies.

Categories:
166 Views

This is the dataset used in the paper "Application of improved lightweight network and Choquet fuzzy ensemble technology for soybean disease identification". This data set contains 6 types of soybean disease leaves collected from Xiangyang Farm, Nengjiang Farm and Jiusan Farm of Northeast Agricultural University in Heilongjiang Province from early June to late September 2019. All images are collected in natural scenes. A total of 1620 disease images of soybean leaves were collected.

Categories:
146 Views

This is the dataset used in the paper "Application of improved lightweight network and Choquet fuzzy ensemble technology for soybean disease identification". This data set contains 6 types of soybean disease leaves collected from Xiangyang Farm, Nengjiang Farm and Jiusan Farm of Northeast Agricultural University in Heilongjiang Province from early June to late September 2019. All images are collected in natural scenes. A total of 1620 disease images of soybean leaves were collected.

Categories:
25 Views

This is the dataset used in the paper "Application of improved lightweight network and Choquet fuzzy ensemble technology for soybean disease identification". This data set contains 6 types of soybean disease leaves collected from Xiangyang Farm, Nengjiang Farm and Jiusan Farm of Northeast Agricultural University in Heilongjiang Province from early June to late September 2019. All images are collected in natural scenes. A total of 1620 disease images of soybean leaves were collected.

Categories:
29 Views

This is the dataset used in the paper "Application of improved lightweight network and Choquet fuzzy ensemble technology for soybean disease identification". This data set contains 6 types of soybean disease leaves collected from Xiangyang Farm, Nengjiang Farm and Jiusan Farm of Northeast Agricultural University in Heilongjiang Province from early June to late September 2019. All images are collected in natural scenes. A total of 1620 disease images of soybean leaves were collected.

Categories:
156 Views

Indoor intelligent perception systems have gained significant attention in recent years. However, accurately detecting human presence can be challenging in the presence of nonhuman subjects such as pets, robots, and electrical appliances, limiting the practicality of these systems for widespread use. 
In this data port, we build the first comprehensive WiFi dataset of motion from various sources in real-world contexts. It includes WiFi data of humans, pets, cleaning robots, and fans. 

Categories:
26 Views

This is the dataset used in the paper "Cross-phone calibration for smartphone-based crowdsourced measurement of E-field strength of mobile downlink signals using transfer learning". The dataset is mainly composed of RSRP and E-field strength data collected using smart phones and the Spectrum Analyzer with isotropic antenna. The file contains two subdirectories, one for the raw data after removing the outliers and the other for the preprocessed feature dataset. See the Readme file in the folder for details.

 

Categories:
132 Views

Mashup and API dataset from ProgrammableWeb.

We have segmented and cleaned the data, retaining useful parts for subsequent task calculations. The storage format is in *. csv format. This mainly consists of two parts of data: Mashup and API, which are mainly used for participating in the post order BERT model. It also includes *. pt data that needs to be used for Node2Vec.

Categories:
90 Views

We downloaded the dataset of Hindi Poems from the Website, contains around 2500 poems the downloaded dataset link is: link In the initial phase of our data preprocessing pipeline, we collected text data from a diverse set of HTML files, totaling 2500 documents. These files, constituting a substantial corpus, were meticulously curated for subsequent analysis. To facilitate further investigation, we amalgamated all the extracted text into a consolidated text file, a crucial step in preparing the data for subsequent processing.

Categories:
225 Views

Pages