Abstract: A technique of scoring a query against a document using sequence to sequence neural networks. The technique comprises: receiving a query comprising a plurality of words from a user; performing a search for a document comprising words based on the query; feeding the words of the document as the input of an encoder of a multilayer sequence to sequence converter; generating a plurality of vectors at a decoder of the multilayer sequence to sequence converter, each vector comprising a probability associated with a respective word in the query; looking up in the respective vector each word's probability of being associated with the document; multiplying every word's probability together to determine an overall probability of the query being associated with the document; and returning the document to the user if the overall probability of the query being associated with the document is greater than a threshold value.

Abstract: A computer-implemented technique is described herein for modifying original keyword information to increase the probability that it will match the queries input by users. The technique operates by using a search engine to provide supplemental information that is relevant to the original keyword information. The technique then mines the supplemental information to extract frequently-occurring n-grams. Next, the technique removes n-grams that are considered to represent noise, and then uses a deep-structured machine-learned model to assign score values to the remaining n-grams. Finally, the technique supplements and/or replaces the original keyword information with the highest-scoring n-grams.

Abstract: Images are encoded into multidimensional vectors in a high-dimensional space utilizing an image model and textual content utilizing a text model. At least one of the image model and/or the text model are tuned such that the point within the multidimensional space pointed to by a vector encoded from an image is proximate to, as determined within the context of that multidimensional space, a point pointed to by a vector encoded from correlated textual content. Received images and textual content are encoded into image vectors and text vectors, respectively, and stored in an image graph and text graph, respectively. An input image can then be encoded as an input image vector and utilized to find close vectors in both the image graph and the text graph, thereby enabling an input image to be utilized to search textual content without using classifiers to deduce textual content therefrom.

Abstract: A computer-implemented technique is described herein for identifying one or more content items that are relevant to an input linguistic item (e.g., an input query) using a deep-structured neural network, trained based on a corpus of click-through data. The input linguistic item has a collection of input tokens. The deep-structured neural network includes a first part that produces word embeddings associated with the respective input tokens, a second part that generates state vectors that capture context information associated with the input tokens, and a third part which distinguishes important parts of the input linguistic item from less important parts. The second part of the deep-structured neural network can be implemented as a recurrent neural network, such as a bi-directional neural network. The third part of the deep-structured neural network can generate a concept vector by forming a weighted sum of the state vectors.

Abstract: A computer-implemented technique is described herein for providing a digital content item using a generator component. The generator component corresponds to a sequence-to-sequence neural network that is trained using an adversarial generative network (GAN) system. In one approach, the technique involves: receiving a query from a user computing device over a computer network; generating random information; generating a key term using the generator component based on the query and the random information; selecting at least one content item based on the key term; and sending the content item(s) over the computer network to the user computing device.

Abstract: A computer-implemented technique is described herein for shortening an original query into one or more sub-queries. The technique chooses the sub-query(ies) such that they preserve the original intent of the original query. To accomplish this goal, the technique uses graph-based analysis to generate a set of richly descriptive query-context-specific feature values for each sub-query, and then uses those feature values to score the relevance of that sub-query.

Abstract: A technique of scoring a query against a document using sequence to sequence neural networks. The technique comprises: receiving a query comprising a plurality of words from a user; performing a search for a document comprising words based on the query; feeding the words of the document as the input of an encoder of a multilayer sequence to sequence converter; generating a plurality of vectors at a decoder of the multilayer sequence to sequence converter, each vector comprising a probability associated with a respective word in the query; looking up in the respective vector each word's probability of being associated with the document; multiplying every word's probability together to determine an overall probability of the query being associated with the document; and returning the document to the user if the overall probability of the query being associated with the document is greater than a threshold value.

Abstract: The present application describes a system and method for converting a natural language query to a standard query using a sequence-to-sequence neural network. As described herein, when a natural language query is receive, the natural language query is converted to a standard query using a sequence-to-sequence model. In some cases, the sequence-to-sequence model is associated with an attention layer. A search using the standard query is performed and various documents may be returned. The documents that result from the search are scored based, at least in part, on a determined conditional entropy of the document. The conditional entropy is determined using the natural language query and the document.

Abstract: Functionality is described herein for analyzing an input linguistic item, such as a query, in a series of stages. The linguistic item includes one or more candidate items. In a first stage, a brand classifier component determines whether the linguistic item specifies at least one brand, to provide a classifier output result. In a second stage, a tagging component generates a set of tags for at least some of the candidate items in the linguistic item, based, in part, on the classifier output result, to generate a tagging output result. An action-taking component then generates at least one result item based on the tagging output result. Functionality is also described herein for producing the brand classifier component and the tagging component using machine-learning training techniques. The training techniques may include provisions to address the later appearance of new brands that do not appear in a brand dictionary.

Abstract: A computer-implemented technique is described herein for identifying one or more content items that are relevant to an input linguistic item (e.g., an input query) using a deep-structured neural network, trained based on a corpus of click-through data. The input linguistic item has a collection of input tokens. The deep-structured neural network includes a first part that produces word embeddings associated with the respective input tokens, a second part that generates state vectors that capture context information associated with the input tokens, and a third part which distinguishes important parts of the input linguistic item from less important parts. The second part of the deep-structured neural network can be implemented as a recurrent neural network, such as a bi-directional neural network. The third part of the deep-structured neural network can generate a concept vector by forming a weighted sum of the state vectors.

Abstract: A computer-implemented technique is described herein for shortening an original query into one or more sub-queries. The technique chooses the sub-query(ies) such that they preserve the original intent of the original query. To accomplish this goal, the technique uses graph-based analysis to generate a set of richly descriptive query-context-specific feature values for each sub-query, and then uses those feature values to score the relevance of that sub-query.

Abstract: A computer-implemented technique is described herein for modifying original keyword information to increase the probability that it will match the queries input by users. The technique operates by using a search engine to provide supplemental information that is relevant to the original keyword information. The technique then mines the supplemental information to extract frequently-occurring n-grams. Next, the technique removes n-grams that are considered to represent noise, and then uses a deep-structured machine-learned model to assign score values to the remaining n-grams. Finally, the technique supplements and/or replaces the original keyword information with the highest-scoring n-grams.

Abstract: Functionality is described herein for transforming first and second symbolic linguistic items into respective first and second continuous-valued concept vectors, using a deep learning model, such as a convolutional latent semantic model. The model is designed to capture both the local and global linguistic contexts of the linguistic items. The functionality then compares the first concept vector with the second concept vector to produce a similarity measure. More specifically, the similarity measure expresses the closeness between the first and second linguistic items in a high-level semantic space. In one case, the first linguistic item corresponds to a query, and the second linguistic item may correspond to a phrase, or a document, or a keyword, or an ad, etc. In one implementation, the convolutional latent semantic model is produced in a training phase based on click-through data.

Abstract: Functionality is described herein for analyzing an input linguistic item, such as a query, in a series of stages. The linguistic item includes one or more candidate items. In a first stage, a brand classifier component determines whether the linguistic item specifies at least one brand, to provide a classifier output result. In a second stage, a tagging component generates a set of tags for at least some of the candidate items in the linguistic item, based, in part, on the classifier output result, to generate a tagging output result. An action-taking component then generates at least one result item based on the tagging output result. Functionality is also described herein for producing the brand classifier component and the tagging component using machine-learning training techniques. The training techniques may include provisions to address the later appearance of new brands that do not appear in a brand dictionary.

Abstract: Systems and Methods for multi-modal or multimedia image retrieval are provided. Automatic image annotation is achieved based on a probabilistic semantic model in which visual features and textual words are connected via a hidden layer comprising the semantic concepts to be discovered, to explicitly exploit the synergy between the two modalities. The association of visual features and textual words is determined in a Bayesian framework to provide confidence of the association. A hidden concept layer which connects the visual feature(s) and the words is discovered by fitting a generative model to the training image and annotation words. An Expectation-Maximization (EM) based iterative learning procedure determines the conditional probabilities of the visual features and the textual words given a hidden concept class. Based on the discovered hidden concept layer and the corresponding conditional probabilities, the image annotation and the text-to-image retrieval are performed using the Bayesian framework.

Abstract: Functionality is described herein for transforming first and second symbolic linguistic items into respective first and second continuous-valued concept vectors, using a deep learning model, such as a convolutional latent semantic model. The model is designed to capture both the local and global linguistic contexts of the linguistic items. The functionality then compares the first concept vector with the second concept vector to produce a similarity measure. More specifically, the similarity measure expresses the closeness between the first and second linguistic items in a high-level semantic space. In one case, the first linguistic item corresponds to a query, and the second linguistic item may correspond to a phrase, or a document, or a keyword, or an ad, etc. In one implementation, the convolutional latent semantic model is produced in a training phase based on click-through data.

Abstract: Systems and methods for responding to an advertisement request with a personalized advertisement are provided. More particularly, in response to an advertisement request from a requesting computer user, a plurality of candidate advertisements are identified and an advertisement is selected. A plurality of potential modifications for the selected advertisement are then identified. A modification, from the plurality of potential modifications, is selected as a function of a user preference vector associated with the requesting computer user. The user preference vector comprises a plurality of user preference items, each item indicating a likelihood of interaction of the requesting computer user with an advertisement having a modification according to a corresponding preference classification. Modification content corresponding to the selected modification is obtained and the modification content is added the selected advertisement.

Abstract: Methods, computer systems, and computer storage media are provided for evaluating information retrieval (IR) such as search query results (including advertisements) by a machine learning scorer. In an embodiment, a set of features is derived from a query and a machine learning algorithm is applied to construct a linear model of (query, ads) for scoring by maximizing a relevance metric. In an embodiment, the machine learned scorer is adapted for use with WAND algorithm based ad selection.

Abstract: Multimedia features extracted from display advertisements may be integrated into a click prediction model for improving click prediction accuracy. Multimedia features may help capture the attractiveness of ads with similar contents or aesthetics. Numerous multimedia features (in addition to user, advertiser and publisher features) may be utilized for the purposes of improving click prediction in ads with limited or no history.

Abstract: Advertisement (“ad”) campaign forecasting may predict results for the campaign and the ads included in the campaign. The results may include a forecast for impressions, clicks, conversions, and/or interactions with the ads. The forecasting may be utilized for display advertising with non-guaranteed delivery (“NGD”) systems in which a bidding platform allows advertisers to bid for ad impressions, clicks, and/or conversions. Forecasting results may be used by advertisers to manage and optimize campaigns.