Microsoft Technology Licensing, LLC

Carolyn P Johnston

Manuel Jesus Reyes-Gomez

Bernard L Johnston

A separate panel may be used to display business icons near images of business entries, if the GIS does not include a business's front door geolocation. Users may place icons that represent business entities near the entrances to the entity. Also, a concise but extensive display of business listing data (e.g., reviews, summaries, services, hours, etc.) in the display near the geolocation and the presentation of further information upon user actions such as a mouse-overs, may avoid browsing away from the viewing application in order to learn more about the business.

Business data display and position correction in street-side imagery

Texts may be classified by mapping the texts to concept space, and by dividing the concept space based on substantive classes. A concept space containing a diverse set of concepts is defined. One example of a concept space is the set of on-line encyclopedia articles, each of which is an example of a concept. A text is scored for relevance against each concept, and a vector is created containing each of the scores. The vector represents the text's position in concept space. For any given substantive class of texts, the concept space may be divided into regions containing texts that are members/non-members of the class. The dividing boundary may be determined by training a classifier on a set of labeled examples of texts that fall inside and outside the class.

Text classification using concept kernel

The present invention is directed to predicting whether two character strings refer to a same subject. An exemplary embodiment includes using a set of character-string pairs, which have been identified as either matches or nonmatches, to learn a function. The function can then be applied to the two character strings to quantify a likelihood that they refer to the same subject matter. For example, a kernel-based classifier analyzes the set of character-string pairs using a kernel function. Based on the analysis the classifier can generate parameters. The parameters are usable to define a prediction algorithm that when applied to the two character strings generates a prediction value, which suggests whether the two characters are matches, i.e., refer to the same subject matter.

Predicting whether strings identify a same subject

Data may be received from the source and based on training; a confidence level may be determined that a specific element in the data is correctly assigned to a master category given that the source assigned the data to one of the plurality of assigned source categories. If the confidence level meets a threshold, the element may be stored in the assigned master category and if the confidence level does not meet a threshold, the element may be stored for reassignment.

Growing community of inventors

Redmond, WA, United States of America

Carolyn P Johnston