Abstract: The system includes a pre-retrieval predictor which determines which collection to submit the query to with a certain degree of confidence. The query is then submitted to either one collection, or multiple collections in parallel. When the results are returned, they are assessed and if they are deemed adequate they are shown to the user. If they are inadequate, the results from the smaller and larger collections are merged and shown to the user. Only if the predictor failed to send the query to more than one collection and the result is not adequate, the query is sent to other collections and executed in a sequential fashion. Overall, large scale searching can be accomplished much more efficiently with no degradation in the quality of the retrieved results and a small increase in processing cost.

Abstract: Disclosed are methods and apparatus for executing a search query. In accordance with one embodiment, a search query is obtained. The search query is classified into one or more of a plurality of categories. The search query is executed for each of the one or more of the plurality of categories. Search results corresponding to the search query are obtained for each of the one or more of the plurality of categories. The search results are then provided for each of the one or more of the plurality of categories.

Abstract: Disclosed are methods and apparatus for automatically storing and generating bookmarks. In one embodiment, a search query is received. Information identifying a bookmark representing the search query is automatically stored in association with a set of bookmarks. Search results corresponding to the search query are automatically obtained and provided, where the search results identify one or more documents. When one of the documents is selected, a link to the selected one of the documents is automatically stored in association with the bookmark.

Abstract: In one embodiment, a method for estimating search query precision is provided, the method comprising: receiving a search query, wherein the search query contains one or more terms; retrieving documents from a collection based on the search query, wherein the retrieving includes only retrieving documents that contain all the terms of the search query; creating a query language model based on the retrieved documents; calculating a divergence between the query language model and the collection; and estimating search query precision based on the divergence, wherein the higher the divergence the more precise the search query.

Abstract: Techniques are described to increase the diversity or focus of image search results. A user submits an original query to search for images. A server generates a first results set by executing the original query using metadata associated with each image. The server selects, from the first results set, a specified number of results ranked highest and generates a list of terms from the metadata of each of the results selected. The terms may be only the tags of the results. The server generates an updated query using terms in the list that may be weighted based on the frequency of the term in the list or include only a specified number of the highest occurring terms in the list. The server generates a second results set by executing the updated query using metadata associated with each image. The second results set is then stored and displayed to the user.

Abstract: A predictor for determining a degree of relevance between a query rewrite and a search query is provided. The predictor may receive a search query from a user via a terminal and identify a set of candidate query rewrites associated with the search query. The predictor may then extract a set of features from advertisements associated with the query rewrites and the search query and determine a degree of relevance between the advertisements and the search query based on a prediction model. The predictor may then determine the degree of relevance between the rewrites and the search query based on the determined degree of relevance between the advertisements and the search query.

Abstract: A method of creating a word game comprising receiving a seed value from a browser, obtaining from a media database a plurality of words associated with the seed value, creating a word game from at least a subset of the obtained plurality of words, integrating the word game into a browser interpretable document, and, returning the browser interpretable document to the browser. Some embodiments further comprise incorporating into the browser interpretable document an advertisement associated with the seed value and/or at least one of the obtained plurality of words. Also disclosed is a system comprising a gaming server which receives a game request; a media server and media tag database; the gaming server requesting from the media server a set of media tags associated with a game seed value, building a word game using at least a subset of the media tags, and transmitting the word game.

Abstract: Systems and methods for building a prediction model to predict a degree of relevance between digital ads and a search query or webpage content are disclosed. Generally, an indication of relevance is received between a plurality of digital ads and one of a webpage content or a search query. A set of features is extracted from the plurality of digital ads and one of the webpage content or the search query. A prediction model is then built to predict a degree of relevance between the set of candidate digital ads and one of a second webpage content or a second search query, where the prediction model is built based at least one the received indication of relevance and the extracted set of features.

Abstract: Systems and methods for predicting a degree of relevance between a set of candidate digital ads and webpage content are disclosed. Generally, an ad provider receives a digital ad request associated with webpage content. The ad provider identifies a set of candidate digital ads that may be served in response to the digital ad request. A relevance module extracts a set of features from the set of candidate digital ads and the webpage content, and determines a degree of relevance between the set of candidate digital ads and the webpage content based on a prediction model and the extracted set of features. If the relevance module determines the set of candidate digital ads is relevant to the webpage content, the ad provider may serve one or more digital ads from the set of candidate digital ads in response to the received digital ad request.

Abstract: Systems and methods for predicting a degree of relevance between a set of candidate digital ads and a search query are disclosed. Generally, an ad provider receives a digital ad request associated with a search query. The ad provider identifies a set of candidate digital ads that may be served in response to the digital ad request. A relevance module extracts a set of features from the set of candidate digital ads and the search query associated with the digital ad request, and determines a degree of relevance between the set of candidate digital ads and the search query based on a prediction model and the extracted set of features. If the relevance module determines the set of candidate digital ads is relevant to the search query, the ad provider may serve one or more digital ads from the set of candidate digital ads in response to the received digital ad request.

Abstract: The subject matter disclosed herein relates to creating a search query based on content and subject of a web page, for example. In one particular example, such a search query may be established by a selection of one or more keywords in a web page. Consequently, the search query may be affected by a determination of content and/or a subject of the web page.

Abstract: A system for and method for ranking results. The system includes a server configured to receive a query and an advertisement engine configured to receive the query from the server. The advertisement engine ranks advertisements based on various features, including at least one word overlap feature and a correlation feature.

Abstract: A system for optimizing machine-learned ranking functions based on click data. The system determines the weighting for each feature of a plurality of features according to a learning model based on the click data. The system selects an element from a plurality of elements for display on a web page based on the weighting of each feature of the plurality of features. The system may rank the items to form a list on the web page based on the weighted features in order of inferred relevance according to the online learning model.

Abstract: Methods, computer products, and systems for selecting advertisements in response to an internet query are provided. The method provides for receiving an internet query that includes query terms, retrieving and then ranking a first set of advertisements in response to the internet query using a query likelihood model. The method then selects sampling words using pseudo-relevance feedback and translation models, the internet query, and the first set of ad materials obtained using the query likelihood model. The sampling words are chosen from a distribution of words from the words in the first set of ad materials, and the pseudo-relevance feedback model is used to select a word (w) in the distribution of words based on a probability that word w generates query term q(p(q|w)). The translation model is used to calculate the probability p(q|w) based on a translation probability that w translates into q(t(q|w)).

Abstract: An ad matching system that includes an interactive client permits a triggering Web page author to provide feedback on a candidate advertisement for the page. Author feedback is used to rank ads for display on the triggering page. Preferably author feedback is also incorporated into ad clustering and/or ad ranking formulae within the system. Also, author credibility is judged based on author feedback and on click through rates of placed ads.

Abstract: A system for selecting electronic advertisements from an advertisement pool to match the surrounding content is disclosed. To select advertisements, the system takes an approach to content match that takes advantage of machine translation technologies. The system of the present invention implements this goal by means of simple and efficient machine translation features that are extracted from the surrounding context to match with the pool of potential advertisements. Machine translation features used as features for training a machine learning model. In one embodiment, a ranking SVM (Support Vector Machines) trained to identify advertisements relevant to a particular context. The trained machine learning model can then be used to rank advertisements for a particular context by supplying the machine learning model with the machine translation features measures for the advertisements and the surrounding context.

Abstract: Systems and methods for identifying spam hosts are disclosed in which hosts known to the system and initially classified as spam or non-spam by a baseline classifier. Then for each node u in the host graph a new feature is computed. This feature is an aggregate function of the initial classifications produced by the baseline classifier for the neighbors of the node u. The set of neighbors can be defined in many different ways: in-link neighbors, out-link neighbors, bi-directional neighbors, k-hops neighbors, etc. The new feature computed above then is added to the existing set of features, and the baseline classifier is trained again, producing new predictions for each node. The results may then be used in many different ways including to filter search results based on host classifications so that spam hosts are not displayed or displayed last in a results set.

Abstract: Systems and methods for identifying spam hosts are disclosed in which hosts are known to the system and initially classified as spam or non-spam by a baseline classifier. The accuracy of the initial host classifications are then improved by propagating them using a random walk algorithm. The random walk used may be modified in order to obtain a weighted or skewed characterization of the host. The hosts may then be reclassified based on the characterization obtained from the random walk to obtain a final spam/non-spam classification. The final classification may then be used in many different ways including to filter search results based on host classifications so that spam hosts are not displayed or displayed last in a results set.

Abstract: Systems and methods for identifying spam hosts are disclosed in which hosts are known to the system and initially classified as spam or non-spam. Then the hosts are partitioned into clusters based on how each host is linked to other hosts. Each cluster is then analyzed and, depending on the number of spam and non-spam hosts it contains, the cluster may be classified as a spam cluster or a non-spam cluster. The hosts within the cluster may then be reclassified based on the cluster's classification. The results may then be used in many different ways including to filter search results based on host classifications so that spam hosts are not displayed or displayed last in a results set.

Abstract: A system for selecting electronic advertisements from an advertisement pool to match the surrounding content is disclosed. To select advertisements, the system takes an approach to content match that focuses on capturing subtler linguistic associations between the surrounding content and the content of the advertisement. The system of the present invention implements this goal by means of simple and efficient semantic association measures dealing with lexical collocations such as conventional multi-word expressions like “big brother” or “strong tea”. The semantic association measures are used as features for training a machine learning model. In one embodiment, a ranking SVM (Support Vector Machines) trained to identify advertisements relevant to a particular context. The trained machine learning model can then be used to rank advertisements for a particular context by supplying the machine learning model with the semantic association measures for the advertisements and the surrounding context.