Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Random Sampling Method01:09

Random Sampling Method

10.9K
Sampling is a technique to select a portion (or subset) of the larger population and study that portion (the sample) to gain information about the population. Data are the result of sampling from a population. The sampling method ensures that samples are drawn without bias and accurately represent the population. Because measuring the entire population in a study is not practical, researchers use samples to represent the population of interest. Among the various sampling methods used by...
10.9K
Nonconscious Mimicry01:13

Nonconscious Mimicry

4.5K
Nonconscious mimicry occurs when individuals alter their mannerisms to match the behaviors and expressions of those nearby, without intention.
4.5K
Cluster Sampling Method01:20

Cluster Sampling Method

11.6K
Appropriate sampling methods ensure that samples are drawn without bias and accurately represent the population. Because measuring the entire population in a study is not practical, researchers use samples to represent the population of interest.
To choose a cluster sample, divide the population into clusters (groups) and then randomly select some of the clusters. All the members from these clusters are in the cluster sample. For example, if you randomly sample four departments from your...
11.6K
Per-Unit Sequence Models01:26

Per-Unit Sequence Models

66
An ideal Y-Y transformer, grounded through neutral impedances, displays per-unit sequence networks akin to those of a single-phase ideal transformer when subjected to balanced positive- or negative-sequence currents. These currents do not produce neutral currents, and their associated voltage drops.
Zero-sequence currents, which are identical in magnitude and phase, generate a neutral current, resulting in voltage drops across the neutral impedance and the low-voltage winding. If the...
66
Modeling and Similitude01:12

Modeling and Similitude

178
Scaled modeling is a fundamental technique in engineering, enabling the study of large and complex systems by creating smaller, manageable replicas that recreate critical characteristics of the original. In hydrology and civil infrastructure, for example, scaled models of dams help analyze water flow, turbulence, and pressure. This method allows for accurate predictions of real-world behavior within a controlled environment, significantly reducing the cost and time involved in full-scale...
178
Stratified Sampling Method01:16

Stratified Sampling Method

11.7K
Sampling is a technique to select a portion (or subset) of the larger population and study that portion (the sample) to gain information about the population. The sampling method ensures that samples are drawn without bias and accurately represent the population. Because measuring the entire population in a study is not practical, researchers use samples to represent the population of interest.
To choose a stratified sample, divide the population into groups called strata and then take a...
11.7K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A global multidimensional analysis of the chimeric antigen receptor T-cell therapy clinical trial landscape and development trends.

Biomarker research·2026
Same author

Freeform gradient-index media represented by rotation-invariant orthogonal polynomials in the unit ball for asymmetric optical design.

Optics express·2026
Same author

Positive emotions and chronic stress in lung cancer patients undergoing immunotherapy: Mediation by perceived life threat and coping strategies.

Journal of psychosomatic research·2026
Same author

Chinese Population-Wide Lipid Trajectories From Age 20 to 85: Sex- and Adiposity-Specific Patterns.

JACC. Asia·2026
Same author

M2 macrophage related genes predict prognosis and drug response in prostate cancer.

Discover oncology·2026
Same author

Routine inflammatory indices modify clonal hematopoiesis-related prognostic risk in patients undergoing percutaneous coronary intervention: a real-world cohort study.

BMC medicine·2026

Related Experiment Video

Updated: May 22, 2025

Perceptual and Category Processing of the Uncanny Valley Hypothesis' Dimension of Human Likeness: Some Methodological Issues
07:34

Perceptual and Category Processing of the Uncanny Valley Hypothesis' Dimension of Human Likeness: Some Methodological Issues

Published on: June 3, 2013

17.3K

Augmenting sparse behavior data for user identity linkage with self-generated by model and mixup-generated samples.

Hongren Huang1, Jianxin Li1, Feihong Lu1

  • 1Beijing Advanced Innovation Center for Big Data and Brain Computing, China; School of Computer Science and Engineering, Beihang University, Beijing, China.

Neural Networks : the Official Journal of the International Neural Network Society
|March 13, 2025
PubMed
Summary
This summary is machine-generated.

This study introduces SGAMDA, a novel data augmentation technique to improve user identity linkage by addressing data sparsity. SGAMDA enhances behavioral data representation, boosting prediction accuracy in recommendation systems.

Keywords:
Data augmentationData sparsityMixup-generatedSelf-generated by modelUser identity linkage

More Related Videos

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
05:48

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

1.4K
Behavioral Tasks for Examining Identity Recognition In Mice
06:58

Behavioral Tasks for Examining Identity Recognition In Mice

Published on: February 7, 2025

358

Related Experiment Videos

Last Updated: May 22, 2025

Perceptual and Category Processing of the Uncanny Valley Hypothesis' Dimension of Human Likeness: Some Methodological Issues
07:34

Perceptual and Category Processing of the Uncanny Valley Hypothesis' Dimension of Human Likeness: Some Methodological Issues

Published on: June 3, 2013

17.3K
Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
05:48

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

1.4K
Behavioral Tasks for Examining Identity Recognition In Mice
06:58

Behavioral Tasks for Examining Identity Recognition In Mice

Published on: February 7, 2025

358

Area of Science:

  • Computer Science
  • Data Science
  • Machine Learning

Background:

  • User identity linkage is crucial for recommendation systems, relying on user-generated behavioral data.
  • Data sparsity, including insufficient user behavior and low-frequency items, poses significant challenges to accurate user modeling.
  • Existing methods struggle with representation errors due to sparse behavioral data.

Purpose of the Study:

  • To propose and evaluate SGAMDA (Self-generated by Model and Mixup-generated Samples-based Data Augmentation) to address data sparsity in user identity linkage.
  • To enhance the accuracy and robustness of user identity linkage models.
  • To improve the representation of user behavior data.

Main Methods:

  • Developed two data augmentation strategies: self-generated samples using Variational Autoencoders and mixup-generated samples.
  • Implemented SGAMDA to generate synthetic training data by decoding representation space samples and mixing behavior data.
  • Categorized user behavior data to guide the application of augmentation strategies based on data volume and low-frequency item proportion.

Main Results:

  • SGAMDA significantly improved prediction accuracy in user identity linkage tasks on the Movies2Books and CDs2Movies datasets.
  • The proposed data augmentation methods effectively enhanced user behavior representation.
  • Demonstrated the efficacy of SGAMDA in alleviating data sparsity issues.

Conclusions:

  • SGAMDA offers a powerful solution for tackling data sparsity in user identity linkage.
  • The approach enhances model performance by improving the quality and quantity of training data.
  • This work contributes to more accurate and reliable user identification in data-driven applications.