Advertisement

Topics

A Most Efficient and Convergent Principal Component Analysis (PCA) Method for Big Data

20:00 EDT 14 May 2019 | NIH

Big data usually means big sample size with many outliers, in which traditional scalable L2-norm principal component analysis (L2-PCA) will fail. Current existing L1-norm PCA (L1-PCA) methods can improve robustness over outliers, however, its scalability is usually limited in either sample size or dimension size.  The inventor proposes an online flipping method to solve L1-PCA challenges, which is not only convergent asymptotically (or with big data), but also achieves most efficiency in the sense each sample is visited only once to extract one principal component (PC). The proposed PCA also has certain robustness to outliers compared to L2-PCA.

If you need a linear complexity robust PCA solver, please contact us; our method can even solve robust PCA in real-time. This efficient robust PCA algorithm is available for licensing and/or collaborations to explore utility for your application.

IC: 
NIDA
NIH Ref. No.: 
E-080-2019
Advantages: 

Current existing L1-norm PCA (L1-PCA) methods can improve robustness over outliers, however, its scalability is usually limited in either sample size or dimension size. The proposed PCA also has certain robustness to outliers compared to L2-PCA

Applications: 
  • Big data analysis 
  • This approach may be the indicated procedure in the presence of unbalanced outlier contamination
Development Status: 

Basic (Target Identification)

Updated On: 
May 15, 2019
Date Published: 
Wednesday, May 15, 2019
Provider Classifications: 
Publications: 
patent field-type-field-collection field-label-hidden">
Licensing Contacts: 
Lead Inventor: 
Inventor IC: 
NIDA
Inventor Lab URL: 
https://irp.drugabuse.gov/staff-members/xiaowei-song/
LPM FIrst Name: 
John
LPM Last Name: 
Hewes
Inv Is lead: 
LPM Phone: 
240-276-5515
LPM Suffix: 
Ph.D.
LPM Organization: 
NCI - National Cancer Institute
DTDT Classification: 
Research Materials
DTDT Description: 
Research Materials
Publication Link: 
https://www.ncbi.nlm.nih.gov/pubmed/25838374
https://ieeexplore.ieee.org/abstract/document/8692807
Publication Caption: 
25838374
IEEE abstract
Publication Title: 

Rosenberg SA, et al. Adoptive cell transfer as personalized immunotherapy for human cancer. 

Song X, An intuitive and most efficient L1-norm principal component analysis algorithm for big data, the 53rd conference on information sciences and systems, 2019 

Collaboration Sought: 
Yes
Institute or Center: 
Collaboration Opportunity: 

Licensing and research collaboration

E Number Only: 
E-080-2019
Inventor First Name: 
Xiaowei
Inventor Last Name: 
Song

Original Article: A Most Efficient and Convergent Principal Component Analysis (PCA) Method for Big Data

NEXT ARTICLE

More From BioPortfolio on "A Most Efficient and Convergent Principal Component Analysis (PCA) Method for Big Data"

Advertisement
Quick Search
Advertisement
Advertisement

 

Relevant Topics

Collaborations in biotechnology
Commercial and academic collaborations are used throughout the biotechnology and pharmaceutical sector to enhance research and product development. Collaborations can take the form of research and evaluation agreements, licensing, partnerships etc. ...

Food
Food is any substance consumed to provide nutritional support for the body. It is usually of plant or animal origin, and contains essential nutrients, such as carbohydrates, fats, proteins, vitamins, or minerals. The substance is ingested by an organism ...