Abstract: Systems and methods for providing auto-completion options to input characters are presented. In response to receiving input characters, a plurality of items of content (that are non-textual items of content) of a corpus of content are identified. These items of content are clustered into n clusters of content according to similarities among the items of content. From the items of content of each cluster, a descriptive title is determined for the cluster. This descriptive title is an auto-completion option for the cluster. The descriptive titles/auto-completion options are provided in response to receiving the input characters.

Abstract: A building system includes one or more storage devices having instructions stored thereon that, when executed by one or more processors, cause the one or more processors to receive an unstructured user question from a user device of a user and query a graph database based on the unstructured user question to extract context associated with the unstructured user question from contextual information of a building stored by the graph database, wherein the graph database stores the contextual information of the building through nodes and edges between the nodes, wherein the nodes represent equipment, spaces, people, and events associated building and the edges represent relationships between the equipment, spaces, people, and events. The instructions further cause the one or more processors to retrieve data from one or more data sources based on the context and compose a presentation based on the retrieved data.

Abstract: An apparatus and method for generating a process enhancement, the apparatus comprising a memory and a processor configured to receive process data, receive user input, determine a plurality of response modules as a function of the user input, determine a modification target as function of the plurality of response modules, wherein determining the modification target includes calculating an importance score for each response module of the plurality of response modules, ranking each response module of the plurality of response modules as a function of the importance score and determining the modification target as a function of the ranking, identify at least a process modification as a function of the process data and the modification target and generate the process enhancement as a function of the at least a process modification.

Abstract: Example aspects of the present disclosure are directed to systems and methods that enable weakly-supervised learning of instance segmentation by applying a cut-and-paste technique to training of a generator model included in a generative adversarial network. In particular, the present disclosure provides a weakly-supervised approach to object instance segmentation. In some implementations, starting with known or predicted object bounding boxes, a generator model can learn to generate object masks by playing a game of cut-and-paste in an adversarial learning setup.

Abstract: Organizations are constantly flooded with questions, ranging from mundane to the unanswerable. It is therefore respective department that actively looks for automated assistance, especially to alleviate the burden of routine, but time-consuming tasks. The embodiments of the present disclosure provide BiLSTM-Siamese Network based Classifier for identifying target class of queries and providing responses to queries pertaining to the identified target class, which acts as an automated assistant that alleviates burden of answering queries in well-defined domains. Siamese Model (SM) is trained for a epochs, and then the same Base-Network is used to train Classification Model (CM) for b epochs iteratively until best accuracy is observed on validation test, wherein SM ensures it learns which sentences are similar/dissimilar semantically while CM learns to predict target class of every user query. Here a and b are assumed to be hyper parameters and are tuned for best performance on the validation set.

Abstract: Embodiments described herein provide dynamic blocking, a decoding algorithm which enables large-scale pretrained language models to generate high-quality paraphrases in an un-supervised setting. Specifically, in order to obtain an alternative surface form, when the language model emits a token that is present in the source sequence, the language model is prevented from generating the next token that is the same as the subsequent source token in the source sequence at the next time step. In this way, the language model is forced to generate a paraphrased sequence of the input source sequence, but with mostly different wording.

Abstract: Methods, systems, and computer programs are presented for providing a prompt tool with interactive entry. One method includes operations for providing a multimodal prompt tool for entering textual description of an item to be generated, and detecting an input that is one of a special character entered in the textual description or a mouse action requesting assistance. Furthermore, a menu is presented with options for the item to be generated, and a list of products, previously added to a data store, is obtained. Further, the method includes providing the list of products for selection; in response to a selection of a product from the list of products, including text associated with the selected product in the textual description; and detecting submittal of the textual description. Further, the textual description is entered as input to a generative artificial intelligence (GAI) tool, and causing presentation of items generated by the GAI tool.

Abstract: In some examples, a computing device may initiate audible play back of a book in response to an instruction from a user. While the book is being audibly played back, the computing device may determine (e.g., in the background) that a content of a recently played back portion of the book matches one or more interests associated with a user, initiate a search, and receive search results. The computing device may re-rank the search results based on the user's interests, pause play back of the book, and audibly read out the re-ranked search results. If the user asks a question, the computing device may pause play back of the book, perform an additional search, and read out loud the additional search results. The computing device may resume the audible play back of the book after the user provides a resume play back instruction.

Abstract: Systems and methods are provided for performing context-based keyword searching using a search bar. Based on search terms received via the search bar, the system may be configured to provide suggested search parameters to associate with that search term. The suggested search parameters may each include a type of data and/or a filter to associate with the search term (e.g., name, phone number, date of birth, etc.). The one or more suggested search parameters may be identified based on the search term itself, a list of possible types of data or filters, a preliminary search of one or more datasets, a record of one or more previous searches performed, requirements associated with one or more searchable datasets, the format of user input received via the search bar, and/or one or more other factors.

Abstract: The invention provides a comprehensive data transformation system, method and computer program product structured for transformation of data structures to maintain data attribute equivalency in diagnostic databases. In some embodiments, the present invention is configured to determine a first source data structure based on at least received user information. The source data structure is typically comprises a comprising a first coding type. In addition, the present invention is configured to determine one or more source attributes associated with the first source data structure. Moreover, the present invention is configured for transforming the first source data structure to a first target data structure of a second coding type such that one or more target attributes associated with the first target data structure are equivalent to the one or more source attributes.

Abstract: Example apparatus disclosed herein are to access first data entries from a first data source based on a first media identifier, the first data entries associated with first streaming media, respective ones of the first data entries including the first media identifier and corresponding timestamps that indicate when the first streaming media was presented or accessed via a group of media devices. Disclosed example apparatus are also to access second data entries from a second data source based on a keyword or phrase, the second data entries associated with news information or weather information. Disclosed example apparatus are further to align, based on the timestamps, the second data entries with values of a time varying audience of the first streaming media determined based on the first data entries to determine ratings data that correlates changes in the time varying audience with the news information or the weather information.

Abstract: A system and method for customer interaction and experience enhancement which automatically gathers direct and indirect customer communications about products and services, converts them to text where necessary, and analyzes the communications for sentiment and emotional content, and scores and displays the information in a manner conducive to making business decisions based on the customer sentiment and emotion, such as making changes to products or services, troubleshooting customer service interactions, and better marketing.

Abstract: A method includes receiving a first search term and a second search term and using the first search term to search a first system that associates the first search term with a plurality of people to retrieve a respective identifier for each of the plurality of people. The second search term and the plurality of identifiers are used to search a second system that associates the second search term with a second plurality of people. The second system returns a count of the number of people who are in both the first plurality of people and the second plurality of people but is prevented from associating the first search term with a person and from releasing the association between the second search term and the second plurality of people. The count is returned as the number of people who have both the first search term and the second search term.

Abstract: Examples of the present disclosure describe systems and methods for determining authoritative documents based on implicit interlinking and communication signals. In aspects, a search operation may be initiated from one or more applications or services. The search operation may be processed to identify search information, such as one or more content items, content topics, or entities. The identified search information may be used to search one or more data sources for implicit relationships between the search information and content items and/or entities stored by the data sources. The results of the search may be collected and ranked according to one or more criteria. The ranked results may be provided in response to the search operation.

Abstract: The present disclosure relates to a method and an apparatus for object preference prediction, and a computer readable medium. The method includes: acquiring evaluation information indicating preference values of partial users in a user set for partial objects in an object set; acquiring auxiliary information of at least one of the user set and the object set, wherein the auxiliary information indicates an attribute of at least one of a corresponding user in the user set and a corresponding object in the object set; determining a user feature representation and an object feature representation using a matrix decomposition model, based on the evaluation information and the auxiliary information; and determining a preference prediction value of a target user in the user set for a target object in the object set based on the user feature representation and the object feature representation.

Abstract: The described technology is directed towards evaluating client-specific information and possibly other information to determine whether to return less data than is available for a data item in response to a request for graph node data to a data service, and/or return expanded graph node data. When a request for graph data is received, client-specific information is used at least in part by a data service in determining whether to return a virtual graph node comprising a relatively lightweight set of information relative to the full set of information maintained in a main (graph) node. The client-specific information is also used in determining whether to return expanded data beyond that requested, e.g., for caching at the client in anticipation of it likely being needed by the client.

Abstract: Ad-hoc table retrieval, including: Representing each of a plurality of tables as a multi-field text document in which: different modalities of the table are represented as separate fields, and a concatenation of all the modalities is represented as a separate, auxiliary field. Receiving a query. Executing the query on the multi-field text documents, to retrieve a list of preliminarily-ranked candidate tables out of the plurality of tables. Calculating an intrinsic table similarity score for each of the candidate tables, based on the query and the auxiliary field. Calculating an extrinsic table similarity score for each of the candidate tables, based on a cluster hypothesis of the candidate tables. Combining: the preliminary rankings, the intrinsic table similarity scores, and the extrinsic table similarity scores, to re-rank the candidate tables.

Abstract: The present invention relates to methods and systems for targeting and retargeting individuals with advertisement spots during television broadcasting. The method and system enable an advertiser for identifying and categorizing a set of viewers or individuals for retargeting advertisement based on parameters such as, but not limited to, interests or preferences of the individuals, past purchases and interactions of the individuals with the advertiser. The method and system further enable the advertiser to segregate the plurality of individuals into subgroups on the basis of information such as, but not limited to, demography, psychographic and behavioral characteristics of the plurality of individuals. The method and system then enable the advertiser to define one or more advertisement spots and corresponding advertisements to be delivered to different sub groups of individuals based on the categorization.

Abstract: An approach is provided that automatically computes a number of feature combinations based on a set of subject matter fields that are selected from a set of subject matter fields and further based on qualifiers that correspond to the selected set of subject matter fields. A model in an artificial intelligence (AI) system is then trained using the computed plurality of feature combinations.

Abstract: Embodiments are disclosed for a method for identifying large database transactions. The method includes generating a token marker sequence of a database transaction. The token marker sequence includes multiple token markers. The token markers include a token of the database transaction and a position corresponding to the token. The method further includes sorting the token markers based on a probability that the token occurs in a stream of database transactions. Additionally, the method includes reducing a size of the token marker sequence based on a predetermined threshold.

Abstract: Methods and systems for incorporating physical documents into a document review workflow involving electronic documents. One or more embodiments detect a presence of a physical document within a field of view of an AR device and map the physical document to an existing electronic document based on visual features of the physical document. Additionally, one or more embodiments determine at least one difference between the physical document and the electronic document and create, for the physical document and the electronic document a shared state mapping including the difference(s). One or more embodiments then apply the difference to the physical document or the electronic document by displaying the difference(s) in an AR layer within the field of view of the AR device or storing the difference(s) in the electronic document.

Abstract: Technologies for attestation techniques, systems, and methods for determining an individual's movement are provided. The individual's movement is determined by passive indoor positioning technology using channel state information (CSI). A first set of data associated with an impulse or frequency response determined using channel state information (CSI) may be received from a Wireless access point located in proximity to a monitored space. A second set of data may also be received from an Internet of Things (IoT) device located in proximity to the monitored space. The first set of data may be compared to the second set of data. A profile database may be updated to replace the first set of data with the second set of data when the first and second sets of data are different.

Abstract: Systems and methods for dynamic computer aided innovation via multidimensional complementary difference recommendation and exploration are disclosed including categorizing a first and second data element in a database with a first attribute and second attribute, respectively, of a first dimension, a dimension being an aspect of a situation, problem, or thing. The first and second data elements are categorized with a first attribute and a second attribute of a second dimension, the second dimension being different from the first dimension. Analyzing the first and second attribute of the first dimension and the first and second attribute of the second dimension to determine a ratio of similarity and dissimilarity; calculating a composite score of the ratio of the first dimension and the ratio of the second dimension; and generating and storing a link between the first and second data element when the composite score is within numerical limits.

Abstract: There are provided a font creation apparatus, a font creation method, and a font creation program capable of generating, from small-number character images having a desired-to-be-imitated style, a complete font set for any language having the same style as the character images. A feature amount extraction unit (40) receives a character image (32) of a first font having a desired-to-be-imitated style and extracts a first feature amount of the first font of the character image (32). An estimation unit (42) estimates a transformation parameter between the extracted first feature amount and a second feature amount of a reference second font (34). A feature amount generation unit (44) generates a fourth feature amount of a second font set to be created by transforming a third feature amount of a complete reference font set (36) based on the estimated transformation parameter.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training an image embedding model. In one aspect, a method comprises: obtaining training data comprising a plurality of training examples, wherein each training example comprises: an image pair comprising a first image and a second image; and selection data indicating one or more of: (i) a co-click rate of the image pair, and (ii) a similar-image click rate of the image pair; and using the training data to train an image embedding model having a plurality of image embedding model parameters.

Abstract: A method for performing an Internet search is provided. The method includes: receiving a textual input that includes at least one search term; using the at least one search term to identify a plurality of Internet-accessible information items that relate to the at least one search term; assigning a respective relevance score to each item of the identified plurality of Internet-accessible information items, such that each respective relevance score is based on a degree of closeness of a textual match between the corresponding information item and the at least one search term; and outputting an ordered list of results, the order of which is based on the relevance scores.

Abstract: A visual dialogue model receives image input and text input that includes a dialogue history between the model and a current utterance by a human user. The model generates a unified contextualized representation using a transformer encoder network, in which the unified contextualized representation includes a token level encoding of the image input and text input. The model generates an encoded visual dialogue input from the unified contextualized representation using visual dialogue encoding layers. The encoded visual dialogue input includes a position level encoding and a segment type encoding. The model generates an answer prediction from the encoded visual dialogue input using a first self-attention mask associated with discriminative settings or a second self-attention mask associated with generative settings. Dense annotation fine tuning may be performed to increase accuracy of the answer prediction. The model provides the answer prediction as a response to the current utterance of the human user.

Abstract: Techniques described herein perform implement a query using a synthetic time series (STS), for example applying an STS on time series data to obtain an answer to the query. In an embodiment, a method receives a times series and query relating to the time series. The method translates the query into a synthetic time series. The method executes a reduce operation on the STS and the time series. The method then returns the result of the operation as an answer to the query. A result of the reduce operation is returned as an answer to the query.

Abstract: A computer based system and method for assigning queries to topics and/or visualizing or analyzing query coverage may include, using a computer processor, searching, using a set of queries, over a set of text documents, to produce for each query a set of search results for the query. Each search result may include a subset of text from a text document of the set of text documents. For each query, a query vector may be calculated based on the set of search results for the query, and for each of a set of topics describing the set of text documents, a topic vector may be calculated. A report or visualization may be generated including the set of queries and the set of topics using the topic vectors and the query vectors.

Abstract: A computing device receives a message including a request for a recommendation. A representation of a hypothetical ideal recommendation to provide in response to the message is determined based on the message content. Data regarding entities that are potential recommendations are retrieved from a data store, the data regarding each entity including a representation of the entity (e.g., a vector) derived from factual information about the entity and opinions of other users of the entity. Ranking scores are determined for at least a subset of the entities based on the difference between the entity representations and the representation of the hypothetical ideal recommendation. An entity to recommend is selected based on the ranking scores and a reply to the message is sent that identifies the selected entity.

Abstract: Methods and devices for assessing the quality of user-generated content are described. In one embodiment, a method is disclosed for measuring the quality of a user-generated answer to a question by combining various factors, including question-answer surface word vector similarity, question-answer explicit semantic analysis vector similarity, answer-answer explicit sematic analysis vector similarity, query performance predictor, sentiment analysis, textual analysis of the answer, and reputation of the answerer. The method uses a learning procedure to determine the best algorithm for measuring the overall quality of the answer based on these factors.

Abstract: Disclosed herein are embodiments of systems, methods, and products comprises a server, which identifies optimized potential suppliers based on the client's request. The request comprises search specification and preselected bidders. The server extracts a first set of keywords from the search specification and finds web pages of potential suppliers based on the first set of keywords. The server identifies the websites of the preselected bidders, extracts a second set of keywords from the websites, and finds web pages of more potential suppliers based on the second set of keywords. The server determines the web pages are associated with real suppliers by excluding non-supplier web pages. The server determines a similarity score for each potential supplier by vectorizing keywords extracted from the supplier's web pages. The server determines a risk score for each potential supplier. The server generates a GUI comprising a list of suppliers ranked based on the similarity scores.

Abstract: A system, computer program product, and method are provided to conduct gap probability mapping to predict presence and location of one or more gaps in a corpus. A probabilistic model is formed to represent inter-corpora associations of objects, which the model leverages to process query submissions. As queries are received and processed, the model creates an adjustment. Subject to evaluation, confidence of the adjustment is evaluated, and a response correlated to the confidence is returned.

Abstract: A method that enables information sharing is described herein. The method includes receiving a query from a first collaborator at a content management system, wherein a relationship between the first collaborator and a second collaborator is maintained by the content management system. The method also includes determining query results in content stored at a second shard belonging to the second collaborator in response to the query, wherein the query results are relevant to a search based on the received query. Additionally, the method includes releasing the query results from the second shard in response to an indication from the second collaborator, and returning the released query results to the first collaborator.

Abstract: A method for providing search suggestions, a device, and a computer-readable storage medium, are provided according to the present disclosure. The method for providing search suggestions includes: obtaining an input search term in a search box of a search interface and determining a first group of suggested search terms associated with the input search term. The method further includes: displaying the first group of suggested search terms in the search interface, and in response to a selection for one or more first suggested search terms from the first group of suggested search terms, searching by taking the input search term and the one or more first suggested search terms as search keywords.

Abstract: A method and an apparatus for outputting information are provided. The method includes: acquiring an info-news data set from a target info-news website; determining a field to which each piece of info-news data in the info-news data set belongs; performing, for each field, a cluster analysis on a piece of info-news data belonging to the field in the info-news data set, to obtain a number of the piece of info-news data and a cluster center in each cluster; and using, for each cluster of each field, the cluster center in the cluster as hot info-news of the field, and using the number of the piece of the information belonging to the cluster as an info-news heat degree of the hot info-news and outputting the info-news heat degree.

Abstract: Disclosed herein are efficient systems and methods for verification of user information in a timely manner. A background computer process for verification of the user information using a third-party authentication service is initiated. The third-party authentication service performs the verification of the information submitted by the user in response to some of the questions within the questionnaire at the same time as the user is replying to remaining questions within the electronic questionnaire. Accordingly, by the time the user is finished replying to all questions within the electronic questionnaire, the previously submitted information by the user is verified and a profile of the user is authenticated.

Abstract: Embodiments provide a computer implemented method for generating a domain-specific thesaurus on a cognitive system, comprising: receiving data of the domain-specific corpus and a plurality of terms of interest from a user; splitting the data of the domain-specific corpus into a plurality of sentences using natural language processing techniques; for each term in the plurality of terms of interest, retrieving a plurality of candidate sentences containing a corresponding term, from the plurality of sentences; for each candidate sentence, providing a list of synonyms of the corresponding term, wherein the synonyms are contextual alternatives in the corresponding candidate sentence; for each term in the plurality of terms of interest, tracking a frequency of each synonym, and forming a frequency map including all the synonyms of a corresponding term and the frequency of each synonym; and generating a domain-specific thesaurus based on a combination of all the synonyms in the frequency map.

Abstract: A method includes generating vertical-specific (VS) records from data sources. Each VS record includes a vertical identifier and a set of VS data fields. The method further comprises generating, for each VS record, an entity partial (EP) record that includes EP data fields populated from the VS data fields. The EP data fields include an entity ID that indicates an entity for the EP record, a source data field that identifies a data source, and an EP searchable data field including data that is descriptive of the entity. The method further comprises generating a search record for each entity ID by combining data from EP records. The data from the EP records is combined based on the source data included in the EP records. Each search record includes a search record searchable data field that includes data from one or more of the EP searchable data fields.

Abstract: Systems, methods, and non-transitory computer readable media configured to determine a value associated with at least one signal indicative of objectionable material in a content item. The value associated with the at least one signal indicative of objectionable material can be compared with a threshold value associated with the at least one signal. A demotion value can be determined in response to satisfaction of the threshold value associated with the at least one signal.

Abstract: Techniques are described herein for determining an information gain score for one or more documents of interest to the user and present information from the documents based on the information gain score. An information gain score for a given document is indicative of additional information that is included in the document beyond information contained in documents that were previously viewed by the user. In some implementations, the information gain score may be determined for one or more documents by applying data from the documents across a machine learning model to generate an information gain score. Based on the information gain scores of a set of documents, the documents can be provided to the user in a manner that reflects the likely information gain that can be attained by the user if the user were to view the documents.

Abstract: The disclosure provides a method and apparatus for processing search data. For a historical search query that includes a knowledge requirement, the disclosure mines entity information for the historical search query and uses that as an answer recommended to users. Thus, the accuracy of entity information recommended to users is improved, and the current problem of poor search results for a historical search query that includes a knowledge requirement is solved.

Abstract: In various embodiments, methods and systems for implementing a semantic signature system are provided. A semantic signature system provides a machine trained semantic representation (i.e., a semantic signature) of the context of a word, synonyms of the word, and weak and strong relationship of the word with other words. The semantic signature can be utilized to facilitate labeling a word that is ambiguous or previously unknown. In practice, the label can be used to more accurately categorize the word for later retrieval by a search or to more accurately provide search results for a search query that includes the word.

Abstract: For query auto completion (QAC) in an eCommerce platform with constantly changing product supply and demand, a Query Mover's Distance (QMD) framework for ranking queries is used, which formulates QAC as an optimal transport problem balancing product demand and supply.

Abstract: The disclosed embodiments provide an augmented reality system. The augmented reality system includes a camera that captures a set of images of one or more physical documents. The augmented reality system also includes a processing apparatus that enables use of the augmented reality system by a user in performing financial management associated with data in the one or more physical documents.

Abstract: The present disclosure provides a method, an apparatus, and a computing device for determining dissemination heat degree of information. The method for determining dissemination heat degree of information includes analyzing N levels of sharing data of specific network information to determine parameters related to each sharing level of the specific network information, where N is a natural number greater than 1; and calculating a dissemination heat degree S of the specific network information based on the parameters related to the sharing of the specific network information. The dissemination heat degree obtained according to the disclosed method, apparatus, and computing device can be used to predict the popular level of the specific network information during the subsequent dissemination process.

Abstract: A method of generating search suggestions includes receiving an indication of a current location of the user device. After a user accesses a search function on the user device and before the user submits a search request, the method includes determining search results associated with locations in proximity to the current location of the user device, using query log data indicating selections of past search results by users after presentation of the past search results in response to respective past queries. The method also includes determining relative positioning of the search results based on the number of users that selected each search result, determining categories associated with the search results, grouping the search results by the determined categories; and sending the grouped search results and the one or more categories associated with the search results to the user device for display according to the determined relative positioning.

Abstract: Disclosed are systems and methods for improving interactions with and between computers in content providing, searching and/or hosting systems supported by or configured with devices, servers and/or platforms. The disclosed systems and methods provide a novel framework that automatically generates and recommends Internet memes for inclusion within electronically communicated messages. As a user is drafting a message, the input text of the drafted message is analyzed and a set of Internet (or visual) memes are compiled and presented to the user within the drafting interface. Upon selection of at least one of the memes, the message is modified by automatically removing the text from the message and replacing the now removed text with the selected meme.

Abstract: Various examples are directed to systems and methods for identifying database records in a database table. A database management system receives a search request comprising a first set of strings associated with a first column of the database table and a second set of strings associated with a second column of the database table. The database management system selects a set of first column keywords using the first set of strings and executes a first tier query at the database table. Responsive to determining that no database record returned by the first tier query has a relevance score greater than a threshold value, the database management system executes a second tier query at the database table.

Abstract: Disclosed herein are systems, methods, and apparatuses that can provide, for example, dynamic page feed recognition for providing zero-configuration analysis of web pages and tailored advertisement provision to website visitors. The systems, methods, and apparatuses can be used to analyze inventory feed. Based on analysis of user interaction data on the web pages, one or more matching methodologies can be used to determine matching data between the visited web pages and the inventory feed.