Microsoft’s “Google Base” is Code Named Fremont & MS Receives Patent For Semi-Auto Annotation of Multimedia Objects

Two items from Redmond today, Microsoft with a classified ads listing service in the works pegged as a rival to Google Base and Microsoft getting a patent on semi-automatic annotation of multimedia objects.

One of the pluses of being “first” is that from then on your service is compared to what may others might have in the works. That’s just the case in this eWeek article by Ben Charny titled: Microsoft Testing Its Own ‘Google Base’.

Microsoft Corp. said it is readying an online marketplace, code-named Fremont, which is apparently in response to a similar feature that rival Google Inc. introduced a few weeks ago.

Charny points out that a is up but can only be accessed and used by MS employees. Michael Arrington has a bit more on TechCrunch. He says to look for a public test in the next few weeks. Before TechCrunch, Greg Sterling posted about the system being an online classifieds move. More in Coming Soon: Windows Live Classifieds.

And while we’re reporting on news from Redmond…

The US Patent and Trademark Office awarded a patent (not a patent app) to Microsoft today titled: Semi-automatic annotation of multimedia objects. It was first filed for in 2000 and is an interesting read, as patents go.

From the abstract:

A multimedia object retrieval and annotation system integrates an annotation process with object retrieval and relevance feedback processes. The annotation process annotates multimedia objects, such as digital images, with semantically relevant keywords. The annotation process is performed in background, hidden from the user, as the user conducts normal searches. The annotation process is “semi-automatic” in that it utilizes both keyword-based information retrieval and content-based image retrieval techniques to automatically search for multimedia objects, and then encourages users to provide feedback on the retrieved objects. The user identifies objects as either relevant or irrelevant to the query keywords and based on this feedback, the system automatically annotates the objects with semantically relevant keywords and/or updates associations between the keywords and objects. As the retrieval-feedback-annotation cycle is repeated, the annotation coverage and accuracy of future searches continues to improve.

And from the summary:

The user interface allows the user to identify multimedia objects that are more relevant to the query, as well as objects that are less or not relevant. The system monitors the user feedback using a combination of feature-based relevance feedback and semantic-based relevance feedback…During the retrieval-feedback-annotation cycle, the system adjusts the weights according to the user feedback, thereby strengthening associations between keywords and objects identified as more relevant and weakening the associations between keywords and objects identified as less relevant. If the association becomes sufficiently weak, the system removes the keyword from the multimedia object. Accordingly, the semi-automatic annotation process captures the efficiency of automatic annotation and the accuracy of manual annotation. As the retrieval-feedback-annotation cycle is repeated, both annotation coverage and annotation quality of the object database is improved.

For more on Content-Based Image Retrieval, this post might be of interest.


Microsoft tries classifieds
from the Seattle Post-Intelligencer (spotted

Greg Linden) has nice details on the project with comments from

Related reading

Search engine results: The ten year evolution
Five ways PPC customer support can help SMBs
#GoogleDoBetter The latest on internal issues at Google and Alphabet
Google Sandbox Is it still affecting new sites in 2019