Reaching the right people

Imagine this situation: A company has hundreds (maybe thousands) of employees. All of them have their own skills and areas of expertise. There is probably lots of overlap, however any one person will not know everyone in the larger group who has particular skill sets. It someone is working on a project and needs assistance to overcome some technical hurdle, it could be very helpful, if they could communicate with those people who also have experience in that area. Those people might be located in entirely different parts of the company.

Semantic email addressing [PDF] aims to solve this problem:

Email addresses are a means to an end. The goal is usually not to send an email to a particular address, but to a particular person. You want to say hello to your friend Steve or send a message to the VP of marketing at Microsoft or to the head caterer for your wedding. Ideally, you could send a message to a person just by entering his or her name, position, or some other descriptive attribute. If a person’s email address changes, the email system should send to the new address automatically. If the person matching a description differs over time, the email system should send to the person currently matching that description.

In the given example, the user would be able to get answers to his or her questions by reaching out to the people with the fitting skill sets without previously having known those people: The email system can decide, who the most appropriate receivers of the messages are.

I cannot help thinking that Aardvark was at least a little inspired by the ideas behind semantic email addressing. Their process is simple: Users send in questions (using email, twitter, IM, etc.), Aarvark routes the question to another user is (hopefully) qualified to answer it and the user will eventually receive a response, often just a few minutes later. In this social search approach, Aardvark accomplishes the job of finding information by finding the right people who can provide it. The service has received very good press and was recently acquired by google.

Twitter seems like it might be a good platform for this problem area. If someone has a public twitter feed, they are essentially broadcasting their updates to the open stream and anyone can see them. It is probably safe to assume, they are at least open to the idea of talking to strangers/responding to messages from people they do not already know.

How could one go about finding the best people to message though? One method is certainly to search the message stream for specific keywords and basically manually look for people who might be active in areas of interest. You can also search in and add yourself to one of the many directories that are being developed.

But, if I simply need to talk to someone and ask them “May I ask you a question about XYZ?” then clearly, a) broadcasting my question hoping that someone will answer could be very inefficient and b) first researching who the best person might be for my question(s) puts all the burden on me.

What if the user could simply send out the question and the system would ensure that the most appropriate people see it?

The basic idea here is this: The user submits the question (along with a set of keywords) to his or her software. The software has analyzed other users’ message streams, extracted keywords, etc. and generated a knowledge base. If the query can be confidently matched to another user, a message is generated and send to that user. The message will be visible to that user as a regular name mention and they can choose whether to engage in that conversation.

Some of the obvious challenges:

• Generating of meaningful keywords/subject areas based on a person’s message stream.
• Successful matching of queries with users.
• Establishing an effective communication protocol that does not easily lend itself to abuse, i.e. spam.

A lot of web-based social networks are great at helping you connect with people you already know. Twitter makes it easy to connect with new people. The outlined approach (or a variation thereof) might be a good way of further supporting creation of those new connections, based on areas of interest.

Peter Norvig on Innovation in Search and Artificial Intelligence

Peter Norvig gave this presentation at Citris on September 2. He emphasizes (with several recent examples), how the usage and availability of large data models and increased computing power improves problem solving approaches.

A lot of interesting subjects are covered in the presentation. Here are references to projects or papers that are mentioned:

• Seam Carving for Content-Aware Image Resizing [PDF] by Shai Avidan and Ariel Shamir presents a smarter method of image resizing. Speed of processing of modern computers greatly helped with the development of this algorithm.
• Scene Completion Using Millions of Photographs by James Hays and Alexei Efros is only possible and successful because of its large data sets.
• The More Data vs Better Algorithms slide is from Michele Banko’s and Eric Brill’s 2001 paper Mitigating the paucity-of-data problem: exploring the effect of training corpus size on classifier performance for natural language processing.
• Canonical image selection from the web by Shumeet Baluja, Yushi Jing and Henry Rowley compares low level features of image result matches for given queries to rank the images.
• Learning people annotation from the web via consistency learning by Jay Yagnik and Atiq Islam uses Eigenface representations and large collections of images to annotate them.
• Audiovisual Celebrity Recognition in Unconstrained Web Videos [PDF] by Mehmet Sargin, Hrishikesh Aradhye, Pedro Moreno and Ming Zhao uses both face and speech recognition to detect celebrities in videos.
• How to Write a Spelling Corrector by Peter Norvig provides spell checking in just over 20 lines of Python.
• Google made an n-gram corpus publicly available at Linguistic Data Consortium.
• Google’s Flu Trends is based on search data.
• Google Sets allows generating of sets of expressions similar to an initial set of expressions.

Also discussed: Text segmentation, statistical machine translation, MapReduce, Web bias and more.

visual search: like.com

Like.com:

Like.com is the first true visual search engine, where the contents of photos are used to search and retrieve similar items.

We believe that there are literally millions of items that are difficult to describe via text-based search and where individual tastes are all over the map — think of your favorite pair of earrings or shoes and what an ordeal it can be to find something new but in a similar style.

We created Like.com to solve the challenge of finding “the perfect you” for a broad range of aesthetic and soft goods.

Like.com utilizes our Likeness Technology™ to create a digital signature that describes the photo’s contents and enables a more accurate search for similar looking items and products.

This is about the most interesting shopping site I have seen lately.

Like.com provides a visual browsing experience. Picking a product results in presentation of other products that share similarities in color and/or shape.

This happens based on the entire product:

As well as based on a selected area of the product (note the small, selected black area on the wristband):

In both examples, the images only depict a small excerpt, of course – the actual number of results is much larger. Anyway – this looks like a very promising application of technology already. Early signs appear good for the company itself as well.

I am looking forward to seeing how they will develop.

Interesting coincidence by the way – just last night I was having a discussion on determining similarities between pictures and their environment, based on theme and/or colors. Visual, contextual matching, if you will.

This is good.