An Interest In:
Web News this Week
- March 29, 2024
- March 28, 2024
- March 27, 2024
- March 26, 2024
- March 25, 2024
- March 24, 2024
- March 23, 2024
Some of Our Sources
- Engadget
- Simplebits
- Team Treehouse
- Smashing Magazine
- Creative Curio
- Fuel Your Creativity
- Stylized Web
- Wal You
- Willems Lab
- Hashedout
Help Webnuz
Referal links:
October 25, 2013 04:00 pm GMT
Lifestyle focused e-commerce company Fab has unveiled a new design for its website in time for the holiday shopping season. The core of Fab.com’s new user interface is its ‘Personalization’ feature, which leverages collaborative filtering technology developed by Fab to make product recommendations more relevant to each shopper’s interests. Fab will also update its iOS and Android apps within the next few days. The company is offering a $10 credit today at this link. Fab.com’s collaborative filtering tech, which it claims is more powerful than that of any other e-commerce site, evaluates the items each visitor has viewed, faved or shared on Facebook, Twitter or Pinterest, past purchases, and activity by other shoppers Fab believes have similar tastes. The site’s recommendation engine takes advantage of data points accumulated from its members, which numbered 14 million in June. Fab and other sites that use collaborative filtering (including Pinterest) constantly try to refine their algorithms so they can provide the most accurate recommendations based on the smallest amount of data. This is important to hook users as early as possible, even if they have clicked on only a few items. Fab’s collaborative filtering technology uses Apache Mahout’s machine learning libraries, which also includes support for user and item based recommendations. The company says its next step is working on singular value decomposition, or the factorization of a real or complex matrix in linear algebra. This helps Fab because collaborative filtering may become slower as the amount of data it has to evaluate increases. Singular value decomposition, on the other hand, offers more scalability and potentially make more accurate predictions using data sets with less dimensionality. Fab’s redesign (which also includes a new user interface that follows the recent trend for flat design among tech companies and larger product photos) comes as the company focuses on expansion in Asia, part of its goal to join a roster of e-commerce companies that are valued at more than $10 billion: Amazon, eBay, Alibaba and Rakuten. But Fab’s growth strategy has not come without its share of headaches. Earlier this year, Fab made the latest in a series of pivots by moving its business model away from flash sales to focus on its “lifestyle shop,” including home goods, which account for 50 percent of sales. The company said it abandoned its flash sales model because it is difficult to scale globally if Fab wants to be
Original Link: http://feedproxy.google.com/~r/Techcrunch/~3/wkwcjCrE9C0/
Fab.com Updates Site To Give Better Recommendations As It Pursues Global Growth
Lifestyle focused e-commerce company Fab has unveiled a new design for its website in time for the holiday shopping season. The core of Fab.com’s new user interface is its ‘Personalization’ feature, which leverages collaborative filtering technology developed by Fab to make product recommendations more relevant to each shopper’s interests. Fab will also update its iOS and Android apps within the next few days. The company is offering a $10 credit today at this link. Fab.com’s collaborative filtering tech, which it claims is more powerful than that of any other e-commerce site, evaluates the items each visitor has viewed, faved or shared on Facebook, Twitter or Pinterest, past purchases, and activity by other shoppers Fab believes have similar tastes. The site’s recommendation engine takes advantage of data points accumulated from its members, which numbered 14 million in June. Fab and other sites that use collaborative filtering (including Pinterest) constantly try to refine their algorithms so they can provide the most accurate recommendations based on the smallest amount of data. This is important to hook users as early as possible, even if they have clicked on only a few items. Fab’s collaborative filtering technology uses Apache Mahout’s machine learning libraries, which also includes support for user and item based recommendations. The company says its next step is working on singular value decomposition, or the factorization of a real or complex matrix in linear algebra. This helps Fab because collaborative filtering may become slower as the amount of data it has to evaluate increases. Singular value decomposition, on the other hand, offers more scalability and potentially make more accurate predictions using data sets with less dimensionality. Fab’s redesign (which also includes a new user interface that follows the recent trend for flat design among tech companies and larger product photos) comes as the company focuses on expansion in Asia, part of its goal to join a roster of e-commerce companies that are valued at more than $10 billion: Amazon, eBay, Alibaba and Rakuten. But Fab’s growth strategy has not come without its share of headaches. Earlier this year, Fab made the latest in a series of pivots by moving its business model away from flash sales to focus on its “lifestyle shop,” including home goods, which account for 50 percent of sales. The company said it abandoned its flash sales model because it is difficult to scale globally if Fab wants to beOriginal Link: http://feedproxy.google.com/~r/Techcrunch/~3/wkwcjCrE9C0/
Share this article:
Tweet
View Full Article
Techcrunch
TechCrunch is a leading technology blog, dedicated to obsessively profiling startups, reviewing new Internet products, and breaking tech news.More About this Source Visit Techcrunch