Ho John Lee's Weblog

Some introductory slides from a panel session at the Social Graph Symposium.

Social Graph Symposium Panel – May 2010 – Presentation Transcript

1. Social Graph Symposium Panel
Ho John Lee | Principal Program Manager | Bing Social Search
2. About me:
Ho John Lee
hojohn . lee @ microsoft . com
twitter.com/hjl
Past: Bing Twitter (v1), SocialQuant, trading, investing/consulting (China, India)
HP Labs, MIT, Stanford, Harvard
Current: Bing Social Search – graph and time series analysis, data mining
Twitter, Facebook, new products, technical planning
3. What can we do by observing social networks?
On the internet, no one knows you’re a dog.
But in social networks, we can tell if you act like a dog, what groups you belong to, and some of your interests
4. How many Twitter users are there?
from a search on twopular, May 2009
5. Graph analysis for relevance and ranking
Spam marketing campaign
(teeth whitening)
Naturally connected community (#smx)
Real time relevance needs data mining to filter and rank based on history
Spammy communities can be highly visible
Social graph, topic/concept graph, and behavior/gesture graphs are all useful tools
6. Information diffusion in the graph
Observed incidence network of retweets in Twitter
Kwak, Lee, et al, What is Twitter, a Social Network or a News Media? WWW2010
Information flow and behaviors form an implicit interaction graph
7. Topic / sentiment range, volume, trend analysis
What is the baseline rate of mentions / sentiment per unit time?
Look for changes in attention flow around a subject, location, topic
Watch for correlated signals from multiple sources
Consider source relevance and authority as well
8. Applying graph analysis
Attention flow vs information flow
Leads to utility functions, cost functions
Variable diffusion rates by actor / network / info type
Predicting interests and affiliations
Content creation follows attention
Self-organized communities of attention
If there’s no content, you can ask for some
Observable propagation of information
9. Clustering and fuzzing properties and identities
* Frequently used terms can identify interests, affinities, latent query intent
* But can potentially be used to identify likely individual users!
* Infochaff – fuzzing out identity, behavior, properties
10. Thank You
Ho John Lee
hojohn . lee @ microsoft . com
twitter.com/hjl

RESEARCH: Insights from the latest social graph studies
Moderator: Eric Siegel – President at Prediction Impact and Conference Chair at Predictive Analytics World
Speakers:
Sharad Goel – Research Scientist at Yahoo
Ho John Lee – Principal Program Manager at Microsoft
DJ Patil – Chief Scientist at LinkedIn
Marc Smith – Chief Social Scientist at Connected Action Consulting Group

These are my links for December 31st through January 17th:

• Khan Academy – The Khan Academy is a not-for-profit organization with the mission of providing a high quality education to anyone, anywhere.

  We have 1000+ videos on YouTube covering everything from basic arithmetic and algebra to differential equations, physics, chemistry, biology and finance which have been recorded by Salman Khan.

• StarCraft AI Competition | Expressive Intelligence Studio – AI bot warfare competition using a hacked API to run StarCraft, will be held at AIIDE2010 in October 2010.
  The competition will use StarCraft Brood War 1.16.1. Bots for StarCraft can be developed using the Broodwar API, which provides hooks into StarCraft and enables the development of custom AI for StarCraft. A C++ interface enables developers to query the current state of the game and issue orders to units. An introduction to the Broodwar API is available here. Instructions for building a bot that communicates with a remote process are available here. There is also a Forum. We encourage submission of bots that make use of advanced AI techniques. Some ideas are:
  * Planning
  * Data Mining
  * Machine Learning
  * Case-Based Reasoning
• Measuring Measures: Learning About Statistical Learning – A "quick start guide" for statistical and machine learning systems, good collection of references.
• Berkowitz et al : The use of formal methods to map, analyze and interpret hawala and terrorist-related alternative remittance systems (2006) – Berkowitz, Steven D., Woodward, Lloyd H., & Woodward, Caitlin. (2006). Use of formal methods to map, analyze and interpret hawala and terrorist-related alternative remittance systems. Originally intended for publication in updating the 1988 volume, eds., Wellman and Berkowitz, Social Structures: A Network Approach (Cambridge University Press). Steve died in November, 2003. See Barry Wellman’s “Steve Berkowitz: A Network Pioneer has passed away,” in Connections 25(2), 2003. It has not been possible to add the updating of references or of the quality of graphics that might have been possible if Berkowitz were alive. An early version of the article appeared in the Proceedings of the Session on Combating Terrorist Networks: Current Research in Social Network Analysis for the New War Fighting Environment. 8th International Command and Control Research and Technology Symposium. National Defense University, Washington, D.C June 17-19, 2003
• SSH Tunneling through web filters | s-anand.net – Step by step tutorial on using Putty and an EC2 instance to set up a private web proxy on demand.
• PyDroid GUI automation toolkit – GitHub – What is Pydroid?

  Pydroid is a simple toolkit for automating and scripting repetitive tasks, especially those involving a GUI, with Python. It includes functions for controlling the mouse and keyboard, finding colors and bitmaps on-screen, as well as displaying cross-platform alerts.
  Why use Pydroid?

  * Testing a GUI application for bugs and edge cases
  o You might think your app is stable, but what happens if you press that button 5000 times?
  * Automating games
  o Writing a script to beat that crappy flash game can be so much more gratifying than spending hours playing it yourself.
  * Freaking out friends and family
  o Well maybe this isn't really a practical use, but…

• Time Series Data Library – More data sets – "This is a collection of about 800 time series drawn from many different fields.Agriculture Chemistry Crime Demography Ecology Finance Health Hydrology Industry Labour Market Macro-Economics Meteorology Micro-Economics Miscellaneous Physics Production Sales Simulated series Sport Transport & Tourism Tree-rings Utilities"
• How informative is Twitter? » SemanticHacker Blog – "We undertook a small study to characterize the different types of messages that can be found on Twitter. We downloaded a sample of tweets over a two-week period using the Twitter streaming API. This resulted in a corpus of 8.9 million messages (”tweets”) posted by 2.6 million unique users. About 2.7 million of these tweets, or 31%, were replies to a tweet posted by another user, while half a million (6%) were retweets. Almost 2 million (22%) of the messages contained a URL."
• Gremlin – a Turing-complete, graph-based programming language – GitHub – Gremlin is a Turing-complete, graph-based programming language developed in Java 1.6+ for key/value-pair multi-relational graphs known as property graphs. Gremlin makes extensive use of the XPath 1.0 language to support complex graph traversals. This language has applications in the areas of graph query, analysis, and manipulation. Connectors exist for the following data management systems:

  * TinkerGraph in-memory graph
  * Neo4j graph database
  * Sesame 2.0 compliant RDF stores
  * MongoDB document database

  The documentation for Gremlin can be found at this location. Finally, please visit TinkerPop for other software products.

• The C Programming Language: 4.10 – by Kernighan & Ritchie & Lovecraft – void Rlyeh
  (int mene[], int wgah, int nagl) {
  int Ia, fhtagn;
  if (wgah>=nagl) return;
  swap (mene,wgah,(wgah+nagl)/2);
  fhtagn = wgah;
  for (Ia=wgah+1; Ia<=nagl; Ia++)
  if (mene[Ia]<mene[wgah])
  swap (mene,++fhtagn,Ia);
  swap (mene,wgah,fhtagn);
  Rlyeh (mene,wgah,fhtagn-1);
  Rlyeh (mene,fhtagn+1,nagl);

  } // PH'NGLUI MGLW'NAFH CTHULHU!

• How to convert email addresses into name, age, ethnicity, sexual orientation – This is so Meta – "Save your email list as a CSV file (just comma separate those email addresses). Upload this file to your facebook account as if you wanted to add them as friends. Voila, facebook will give you all the profiles of all those users (in my test, about 80% of my email lists have facebook profiles). Now, click through each profile, and because of the new default facebook settings, which makes all information public, about 95% of the user info is available for you to harvest."
• Microsoft Security Development Lifecycle (SDL): Tools Repository – A collection of previously internal-only security tools from Microsoft, including anti-xss, fuzz test, fxcop, threat modeling, binscope, now available for free download.
• Analytics X Prize – Home – Forecast the murder rate in Philadelphia – The Analytics X Prize is an ongoing contest to apply analytics, modeling, and statistics to solve the social problems that affect our cities. It combines the fields of statistics, mathematics, and social science to understand the root causes of dysfunction in our neighborhoods. Understanding these relationships and discovering the most highly correlated variables allows us to deploy our limited resources more effectively and target the variables that will have the greatest positive impact on improvement.
• PeteSearch: How to find user information from an email address – FindByEmail code released as open-source. You pass it an email address, and it queries 11 different public APIs to discover what information those services have on the user with that email address.
• Measuring Measures: Beyond PageRank: Learning with Content and Networks – Conclusion: learning based on content and network data is the current state of the art There is a great paper and talk about personalization in Google News they use content for this purpose, and then user click streams to provide personalization, i.e. recommend specific articles within each topical cluster. The issue is content filtering is typically (as we say in research) "way harder." Suppose you have a social graph, a bunch of documents, and you know that some users in the social graph like some documents, and you want to recommend other documents that you think they will like. Using approaches based on Networks, you might consider clustering users based on co-visitaion (they have co-liked some of the documents). This scales great, and it internationalizes great. If you start extracting features from the documents themselves, then what you build for English may not work as well for the Chinese market. In addition, there is far more data in the text than there is in the social graph
• mikemaccana’s python-docx at master – GitHub – MIT-licensed Python library to read/write Microsoft Word docx format files. "The docx module reads and writes Microsoft Office Word 2007 docx files. These are referred to as 'WordML', 'Office Open XML' and 'Open XML' by Microsoft. They can be opened in Microsoft Office 2007, Microsoft Mac Office 2008, OpenOffice.org 2.2, and Apple iWork 08. The module was created when I was looking for a Python support for MS Word .doc files, but could only find various hacks involving COM automation, calling .net or Java, or automating OpenOffice or MS Office."

What’s Different about Real Time and Social Search – HJL Slides For SES Chicago Dec 09

Crossroads of the World at the Beach Bar, Waikiki

As some of you know, I have been exploring a variety of paths forward for SocialQuant, my real time social search and analytics project. My family, friends, and colleagues have given me much support, patience, and advice during this process, which has reached a crossroads, and as Yogi Berra says, “When you come to a fork in the road, take it!”

The rise of Twitter, Facebook, and other social media, combined with web-based applications, smartphones, and cloud computing have all set the stage for new applications and use models based on social discovery, collaboration, and communications, in addition to traditional search. What we’re all calling “real time search” lately isn’t exactly real time, nor is it exactly search, in which you find a definitive/authoritative answer. Much of the opportunity revolves around discovering people, discussions, and events that are relevant to you and bringing it to your attention in a timely, actionable fashion. Information streams from social media are transient, unreliable, and noisy. At the same time, the sheer volume of data can help provide the basis for building better filters. As an added bonus, you can ask questions to people in the social graph itself, and there are numerous examples of communities of interest forming around current events such as Barack Obama’s inauguration, the Iran elections, or even Michael Jackson’s funeral, all of which help surface information content, opinion, and sentiment that were previously inaccessible online. One interesting aspect of real time social media is that it’s not just algorithmic, it’s based on human connections and emotions. So a message  that “feels right” from people you trust can be more relevant than one that is “correct” at times.

The challenge then is in filtering and ranking the massive flow of information in a way that helps direct the user’s limited (and non-expanding) time and attention in a way that’s most valuable to them. With today’s information technology, amazing things are possible with limited resources. I personally have more computing and storage resources than the facility we launched HP’s original photo site with (for millions of dollars), at a fraction of the cost, routinely pushing around datasets of millions of rows on the local development servers. Unfortunately, that’s just the ante to get started on the problem. Running ranking, clustering, and semantic analysis for filtering the ever-growing stream of social media eventually requires web scale computing, even with careful problem selection and data pruning. The bar is also going up every day as the social media user base grows, and as well funded teams make progress on their platforms (+Google).  So very shortly, to be competitive in real time, social search and discovery is going to require access to lots of data and either getting a datacenter or working with someone who has one.

In my case, I have recently chosen the latter path, and will be joining the Microsoft Bing search team, focusing on real time and social search. Microsoft itself has been showing signs of a renaissance, with search relaunching, Windows 7 looking leaner, Azure becoming non-vaporous, more web APIs getting published, core online applications starting to turn up, and a cool Office 2010 video. Even Mini-Microsoft is getting positive recently. And Google is starting to have “bigness” issues.

I look forward to working with Sean Suchter and the Microsoft Bing search team (and likely expanding their carbon footprint) in pursuit of new applications and services as the social media and online application space evolves.

You can follow along on Twitter (@hjl). As always, any and all opinions here are solely mine and do not reflect the position of any past, present, or future employer, partner, or business associate.