Microsoft Technology Licensing, LLC

Google Inc.

Dennis Craig Fetterly

Michael Isard

Ulfar Erlingsson

Yuan Yu

Mihai Budiu

Jin Li

Eugene Jon Shekita

Marc Alexander Najork

Mark Steven Manasse

Cheng Huang

Sergey M Yekhanin

Alexandros Ntoulas

Parikshit Santhan Gopalan

Steve Chien

Robert John Jenkins, Jr

Steffen Rendle

Bor-yiing Su

Steve Shaw-Tang Chien

The present disclosure provides a new scalable coordinate descent (SCD) algorithm and associated system for generalized linear models whose convergence behavior is always the same, regardless of how much SCD is scaled out and regardless of the computing environment. This makes SCD highly robust and enables it to scale to massive datasets on low-cost commodity servers. According to one aspect, by using a natural partitioning of parameters into blocks, updates can be performed in parallel a block at a time without compromising convergence. Experimental results on a real advertising dataset are used to demonstrate SCD's cost effectiveness and scalability.

Robust large-scale machine learning in the cloud

General-purpose distributed data-parallel computing using a high-level language is disclosed. Data parallel portions of a sequential program that is written by a developer in a high-level language are automatically translated into a distributed execution plan. The distributed execution plan is then executed on large compute clusters. Thus, the developer is allowed to write the program using familiar programming constructs in the high level language. Moreover, developers without experience with distributed compute systems are able to take advantage of such systems.

General purpose distributed data parallel computing using a high level language

In some examples, an erasure code can be implemented to provide for fault-tolerant storage of data. Maximally recoverable cloud codes, resilient cloud codes, and robust product codes are examples of different erasure codes that can be implemented to encode and store data. Implementing different erasure codes and different parameters within each erasure code can involve trade-offs between reliability, redundancy, and locality. In some examples, an erasure code can specify placement of the encoded data on machines that are organized into racks.

Local erasure codes for data storage

A method for static ranking of web documents is disclosed. Search engines are typically configured such that search results having a higher PageRank&#xae; score are listed first. A modified scoring technique is provided whereby the score includes a reset vector that is biased toward web pages linked to blogs. This requires identifying web pages as either blogs or non-blogs.

Ranking method using hyperlinks in blogs

Evaluating content includes receiving content, analyzing the content for web spam using a content-based identification technique, and classifying the content according to the analysis. An index of analyzed contents may be created. A system for evaluating content includes a storage device configured to store data and a processor configured to analyze content for web spam using content-based identification techniques.

Using content analysis to detect spam web pages

A machine learning classifier is used to determine whether a web page belongs to a blog, based on a number of characteristics of web pages (e.g., presence of words such as 'permalink', or being hosted on a known blogging site). The classifier may be initially trained using human-judged examples. After classifying web pages as being blog pages, the blog pages may be further identified or categorized as top level blogs based on their URLs, for example.

Identifying a web page as belonging to a blog

Growing community of inventors

Belmont, CA, United States of America

Dennis Craig Fetterly