Abstract: Mechanisms are provided for ingesting natural language textual content. Ingestion of natural language textual content is initiated and an embedded structured data portion within the natural language textual content is identified. A signature of the structured data portion is generated which comprises one or more metadata elements describing the configuration or content of the structured data portion. References to the structured data portion are identified in natural language text portions of the natural language textual content and evaluated based on the signature. An ingestion prioritization plan for ingesting a set of facts associated with a set of elements of the structured data portion is generated based on results of the evaluation. The ingestion prioritization plan is applied to generate the set of facts and store the set of facts in an ingested representation of the natural language textual content.

Abstract: The disclosed embodiments relate to a system that updates a context that facilitates evaluating qualitative search terms for an attribute during query processing. During operation, the system extracts a value for the attribute from each data item in a set of data items. Next, the system updates the context based on the extracted attribute values, wherein the context includes a concept-mapping for one or more qualitative search terms applied to the attribute, and wherein each concept-mapping associates a given attribute value with a numerical compatibility index that indicates a compatibility between the given attribute value and a corresponding qualitative search term.

Abstract: Multiquery optimization is performed in the context of RDF/SPARQL. Heuristic algorithms partition an input batch of queries into groups such that each group of queries can be optimized together. The optimization incorporates an efficient algorithm to discover the common sub-structures of multiple SPARQL queries and an effective cost model to compare candidate execution plans. No assumptions are made about the underlying SPARQL query engine. This provides portability across different RDF stores.

Abstract: Computer implemented methods for executing a database query in a database using an electronic computing device are presented, the method including: causing the electronic computing device to provide at least one data structure including at least one object having at least two properties of different data types; identifying a property of the at least one data structure that includes the smallest unique values in terms of type of data stored in that particular property; executing the database query, including any limiting parameters, configured to retrieve only data from the property including the smallest unique values that are retrievable for a record; retrieving a predefined number of results of the executed database query; and for each set of predefined number of retrieved results, executing a new database access thread that is configured to retrieve data that are present in the records identified with the smallest unique values.

Abstract: A continuous query parallel engine on data streams provides scalability and increases the throughput by the addition of new nodes. The parallel processing can be applied to data stream processing and complex events processing. The continuous query parallel engine receives the query to be deployed and splits the original query into subqueries, obtaining at least one subquery; each subquery is executed in at least in one node. Tuples produced by each operator of each subquery are labeled with timestamps. A load balancer is interposed at the output of each node that executes each one of the instances of the source subquery and an input merger is interposed in each node that executes each one of the instances of a destination subquery. After checks are performed, further load balancers or input managers may be added.

Abstract: Consistent contextual patterns may confirm ownership. Current usage of mobile and smart devices may be compared to historical usage. If a device is being used as historically observed, then ownership of the device may be confirmed. If, however, the current usage fails to coincide with historical usage, new ownership may be inferred.

Abstract: Implementations generally relate to processing similar documents. In some implementations, a method includes receiving a plurality of documents related to e-discovery. The method further includes determining a seed document from the plurality of documents. The method further includes receiving a search request to search at least one selection of text in the seed document. The method further includes identifying other documents from the plurality of documents based on a similarity between text in the other documents and the at least one selection of text in the seed document. The method further includes generating a graphical user interface that includes a similarity panel that provides similarity data between text in the other documents and the at least one selection of text in the seed document.

Abstract: Personalized search or query suggestions associated with one or more persons and/or content items are provided. A suggestion application learns from user behavior within the suggestion application and presents suggestions for allowing the user to search or navigate to one or more people of particular interest or relevance to the user and for allowing the user to search or navigate to one or more content items associated with people and activities of particular interest or relevance to the user. Two types of suggestions are provided to the user. A first type of suggestion involves suggesting one or more people that may be of particular relevance or interest to the querying user. A second type of suggestion includes a textual suggestion comprised of a person (actor) and an associated action.

Abstract: The disclosure generally describes methods, software, and systems, including a method for generating search results. A search query for querying information from an information source, including business objects, is received from a user. Features associated with the user and business objects are identified for which result items are to be boosted using business knowledge associated with the user and information source. For each identified feature, query terms in the received search query applicable to a particular feature are identified. Weights based on the business knowledge are identified for feature-value pairs associated with each query term. A weighted search query is generated that includes weighted query parts that are a function of a respective query term and associated weight. The weighted search query is executed to produce a ranked result set that includes result items ranked according to the weights. The ranked result set is provided.

Abstract: A query processing service processes a query pertaining to a question about a client device. The query processing service identifies a subset of answers, from a plurality of answers, which are determined to be most relevant to the query, based upon relevance data for the plurality of answers. The query processing service provides to a client device answer data that includes the subset of answers. The query processing service receives, from the client device, user selection data that specifies a user selection of a particular answer to the question about the client device. In response to receiving the user selection data, the query processing service updates user selection data maintained by the query processing service to include the user selection of the particular answer to the question about the client device. The query processing device may also revise relevance data based upon the update made to the user selection data.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage medium, for performing a blend polygon search. In one aspect, the method may include receiving one or more search parameters. Additional actions may include obtaining two or more sets of data that each represent a respective geographic area, the two or more sets of data including at least one set of data that is provided by a third party geographic data set provider, defining a particular geographic area to search based at least on the two or more sets of data that each represent a respective geographic area, identifying one or more entities that (i) are indicated as responsive to one or more of the search parameters, and (ii) are associated with the particular geographic area, and providing a representation of the entities for output.

Abstract: Methods and apparatus are described for scoring documents in response, in part, to parameters related to the document, source, and/or cluster score. Methods and apparatus are also described for scoring a cluster in response, in part, to parameters related to documents within the cluster and/or sources corresponding to the documents within the cluster. In one embodiment, the invention may identify the source; detect a plurality of documents published by the source; analyze the plurality of documents with respect to at least one parameter, and determine a source score for the source in response, in part, to the parameter. In another embodiment, the invention may identify a topic; identify a plurality of clusters in response to the topic; analyze at least one parameter corresponding to each of the plurality of clusters; and calculate a cluster score for each of the plurality of clusters in response, in part, to the parameter.

Abstract: In an example, a plurality of previously submitted confidential data values of a first confidential data type retrieved for a slice having one or more attributes. For a confidential data type, one or more submitted confidential data values of the confidential data type from the slice that are considered outliers based on an external data set or internal data set. A confidence score is calculated by multiplying a support score for the confidential data type in the slice by a non-outlier score for the confidential data type in the slice, the support score being equal to n?/(n?+c), where c is a smoothing constant and n? is the number of non-excluded submitted confidential data values of the confidential data type in the slice and the non-outlier score being equal to n?/n, where n is the total number of non-null submitted confidential data value of the confidential data type in the slice.

Abstract: Techniques are disclosed herein for presenting feature images corresponding to visual depictions of features or products matching a particular room type and décor style. The techniques may be used assist homeowners in the design or decoration of their home, thereby improving home improvement results and the user experience. The techniques described herein allow the homeowner or user to adequately select building materials and décor items by themselves because they can rely on other users' experiences, a form of crowdsourcing that promotes satisfactory results. The techniques disclosed herein may provide a crowdsourced starting point based on an iterative process and customization options based on the particular user's own taste and budget.

Abstract: A method for detecting and locating occurrence in a data stream of any complex string belonging to a predefined complex dictionary is disclosed. A complex string may comprise an arbitrary number of interleaving coherent strings and ambiguous strings. The method comprises a first process for transforming the complex dictionary into a simple structure to enable continuously conducting computationally efficient search, and a second process for examining received data in real time using the simple structure. The method may be implemented as an article of manufacture having a processor-readable storage medium having instructions stored thereon for execution by a processor, causing the processor to match examined data to an object complex string belonging to the complex dictionary, where the matching process is based on equality to constituent coherent strings, and congruence to ambiguous strings, of the object complex string.

Abstract: A building management system (BMS) includes building equipment configured to provide raw data samples of one or more data points in the BMS. The BMS further includes a data collector configured to collect raw data samples from the building equipment and generate one or more raw data timeseries comprising a plurality of the raw data samples. The BMS also includes a timeseries processing engine. The timeseries processing engine is configured to identify one or more timeseries processing workflows that apply to the raw data timeseries, each of the workflows comprising a predefined sequence of timeseries processing operation. The timeseries processing engine is further configured to process the raw data timeseries using the identified timeseries processing workflows to generate one or more derived data timeseries. The BMS further includes a timeseries storage interface configured to store the raw data timeseries and the derived data timeseries in a timeseries database.

Abstract: A flexible approach for storing time series data, utilizes multiple representations in order to achieve optimization amongst various dimensions such as covered time period, accuracy, compression model, and storage medium. A first aspect selectively provides different representations for various portions of a time series along the time axis (horizontal). In this horizontal aspect, a first compression model may be applied to store data occurring during one period in a time series, while a different compression model is applied to store data during another period. In a separate (vertical) aspect, time series data over a same time period may be saved in multiple representations using different compression models. While increasing demands on memory, such vertical storage can afford efficient access for specific purposes (i.e., analytical).

Abstract: A database language (e.g. SQL) is extended to define a transient field whose value is derived from data stored in a database. The value of the transient field is calculated by the database engine for consumption as needed by the application layer, for example as part of a particular data model. The value of the transient field may not be materialized as a persistent field in the underlying database table, or may be selectively materialized based upon a heuristic or hint. Certain embodiments may implement the transient field as a basic elaboration on standard SQL utilizing an identifier. Some embodiments may implement the transient field as an entity where SQL has been extended to accommodate concepts of a higher-level Entity Relationship Model (ERM). The use of constants and/or floating values (e.g. a current time) in connection with transient fields, is also disclosed.

Abstract: Data engine integration and data refinement are described. The actions include receiving, by an in-memory engine, a request for data. The actions include transmitting, by the in-memory engine and to a metadata storage unit, data that is associated with the request. The actions include receiving metadata. The actions include determining a first amount of processing to be performed by the in-memory engine and a second amount of processing to be performed by a data engine. The actions include transmitting a request (i) for a first portion of unprocessed data and (ii) to perform a second amount of processing on a second portion of unprocessed data. The actions include receiving the first portion of unprocessed data and the second portion of processed data. The actions include processing the first portion of unprocessed data. The actions include, providing the first portion of processed data and the second portion of processed data.

Abstract: Techniques are disclosed for consuming and integrating with representational state transfer (REST) based services. In one embodiment, a multiple-step extract, transform, load (ETL) job includes one or more REST steps. Each step in the ETL job performs a particular well-defined task, and the REST step in particular constructs REST requests, invokes REST services, and retrieves REST responses. The REST step may be mapped to a vector of inputs such that multiple REST service calls are made for each of the inputs. Further, a reusable connection may be employed so that one connection can be used to make multiple REST service calls. URLs used to make REST service calls may be dynamically constructed based on input data parameters and ETL job parameters. Additionally, a graphical user interface (GUI) allows users to configure aspects of a REST step.

Abstract: A contextual analysis system that extracts data elements from an unstructured text input; determines whether the extracted data elements are relevant to a predetermined context; and determines, for the extracted elements deemed as relevant, whether the information contained in the relevant data elements complies with a guideline.

Abstract: A computer-implemented method renders hierarchical multi-attribute data in a comparative layout on a display device. The method allocates graphical areas for value partitions of the normalized hierarchy of value partitions within a graphical user interface on the display device. The allocating follows the structure and the order of the normalized hierarchy of value partitions, such that the graphical area for a parent value partition contains the graphical area for its children value partitions in a first dimension. For a proportional layout, lengths of the first dimension of the graphical areas for structurally parallel value partitions are equal, and a ratio of lengths of the first dimension of the graphical areas for comparable value partitions equals a ratio of maximal values among structurally parallel value partitions for each corresponding comparable value partition. The method renders a graphical representation of values within the graphical areas of the value partitions.

Abstract: Systems and methods are provided for generating reports reflecting a visualization of, e.g., a dataset on which data analysis has been performed. The data analysis performed on the dataset may be an aggregation of data analysis functionality provided by applications or tools. Dataset visualization is realized in a workspace implemented by a separate application or tool. Potential conflicts regarding updating one or more elements of the dataset are reconciled by implementing a dual operational transform system and method. A global state of the dataset may be determined by a first operational transform system and method, while a local state of the dataset may be determined by a second operational transform system and method. The local state of the dataset may be updated through operational transforms exchanged between the first and second operational transform systems that align the local state of the dataset with the global state of the dataset.

Abstract: Technologies are described for performing replication within a database environment. Where a database transaction is replicated at multiple replica nodes, a replica node is selected as a coordinator replica node for the transaction. The other replica node or nodes are designated as follower replica nodes for the transaction. A follower replica node sends the coordinator replica node a precommit notification when the follower replica node has precommitted the transaction. The coordinator replica node sends the follower replica node a postcommit notification to commit the transaction when the transaction has been precommitted by all of the replica nodes to which the transaction is to be replicated.

Abstract: The present disclosure describes techniques and apparatuses for a hardware-implemented Adelson-Velskii and Landis' (AVL) tree module. In some aspects, commands are received at the AVL tree module that request operations be performed for an AVL tree table stored in memory. Each command is written to one of multiple hardware threads of the AVL tree module that perform AVL tree operations by causing records of the AVL tree table to be read from memory using a single read operation and then written to the AVL tree module, modifying those records without accessing memory, and causing modified records to be written back to memory using a single write operation. Once a command is written to a hardware thread, the hardware thread can perform the corresponding AVL tree operation. After the AVL tree operation is performed, results are returned to the requesting agent to indicate whether the AVL tree operation was performed successfully.

Abstract: An automated system for validating and refining a template warehouse star schema for a customized data source.

Abstract: Media content may be recommended based on feedback weightings. Input data describing the presentation of a media content items in association with content management accounts may be represented as data points. Each data point may identify feedback data for a media content item previously recommended for presentation in association with a content management account. The feedback data may identify a viewer reaction to the recommended media content item. A weighting factor based on the feedback data for the media content item presented in association with the content management account may be applied to produce a recommendation.

Abstract: The disclosed computer-implemented method for aggregating information-asset classifications may include (1) identifying a data collection that includes two or more information assets, (2) identifying a classification for each of the information assets, (3) deriving, based at least in part on the classifications of the information assets, an aggregate classification for the data collection, and (4) associating the aggregate classification with the data collection to enable a data management system to enforce a data management policy based on the aggregate classification. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A method for dynamically clustering data items, the method comprising: receiving a plurality of data items originating from at least two sources, a plurality of distinct metadata details, and data indicative of associations between the data items and the metadata details, wherein each data item is associated with at least one metadata detail indicative of its owner, and wherein at least a first data item originating from a first source and a second data item originating from a second source are related data items associated with at least one shared metadata detail; grading probabilities of relationships between at least one of the data items and at least one of the metadata details; clustering the data items into one or more clusters, based on the calculated probabilities; and, optionally, sharing clusters and meta-clusters between users.

Abstract: In an example, one or more leaf category specific unsupervised statistical language model (SLM) models are trained using sample item listings corresponding to each of one or more leaf categories and structured data about the one or more leaf categories, the training including calculating an expected perplexity and a standard deviation for item listing titles. A perplexity for a title of a particular item listing is calculated and a perplexity deviation signal is generated based on a difference between the perplexity for the title of the particular item listing and the expected perplexity for item listing titles in a leaf category of the particular item listing and based on the standard deviation for item listing titles in the leaf category of the particular item listing. A gradient boosting machine (GBM) fuses the perplexity deviation signal with one or more other signals to generate a miscategorization classification score corresponding to the particular item listing.

Abstract: A multi-level approach for generating suggestions and/or recommendations associated with one or more items is provided. A group of related items may be identified, for example, by generating a collaborative filtering graph or other graph that includes items and connections between the items. Description data associated with items included in the group may be evaluated in order to determine respective keywords associated with the items and/or a list of common keywords representative of the group of items or a subset of the group of items. Based at least in part upon the list of common keywords, at least one suggestion may be generated.

Abstract: A method, a computer system, and a computer program product are provided for calculating a degree of similarity of users. Each user may be a preparer and an answerer of one or more quizzes. The method includes collecting one or more quizzes from one or more users as preparers and providing the collected quizzes to one or more user as answerers. The method further includes collecting one or more answering activities of one or more answerers to the provided quizzes from the one or more answerers, wherein a pair includes a preparer and an answerer. The method additionally includes calculating, for each pair, a degree of similarity of knowledge background, based on the one or more answering activities of the one or more answerers.

Abstract: A system and machine-implemented method for processing a social endorsement for an item. Structured data for a first item is received in association with a first webpage, the structured data for the first item defining properties of the first item that are specific to the first item. An indication of a social endorsement for the first item is received in association with the first webpage. In response to receiving the indication of the social endorsement, an item element corresponding to the first item is identified within a data structure. The data structure stores plural item elements, each item element including a description of the item and social endorsement information associated with the item. The social endorsement information associated with the identified item element is updated within the data structure, based on the received indication of the social endorsement for the first item.

Abstract: A data manager determines an appropriate number of clusters for continuous data using unsupervised learning. The data manager selects an appropriate number of clusters based on at least one temporal stability measure between continuous data from at least two time intervals.

Abstract: Embodiments of the present invention disclose a method, a computer program product, and a computer system for identifying information gaps in corpora. A computer receives a document and extracts keywords from the document while filtering trivial keywords. The computer identifies and extracts top keywords detailed by the document using a topic modelling approach before determining whether the extracted top keywords exceed a threshold use frequency. Based on determining that the top keywords exceed a threshold use frequency, determining whether the top keywords have a relation to other entities within the document and, if so, determining whether the top keywords are defined within the document. Based on determining that the top keywords are not defined in the document, adding the top keywords to a list and defining the top keywords.

Abstract: The present teaching relates to searching encrypted data. In one example, a search request is received for encrypted documents. An encrypted query is generated based on the search request. The encrypted query is sent to a server that stores a first encrypted index and a second encrypted index. The first encrypted index maps encrypted keywords to full blocks each of which has a same size and is fully filled with encrypted document identities (IDs). The second encrypted index maps encrypted keywords to partial blocks each of which has the same size and is partially filled with encrypted document IDs. Based on the encrypted query, one or more encrypted document IDs are determined by searching against both the first encrypted index and the second encrypted index. A search result is generated based on the one or more encrypted document IDs. The search result is provided in response to the search request.

Abstract: Aspects and implementations of the present disclosure are directed to methods of and systems for search lift remarketing. In general, in some implementations, a first content item is distributed to client devices and search lift attributable to the first content item is measured by examining subsequent requests received from client devices to which the first content item has been distributed as compared to requests received from similar client devices to which the first content item has not been distributed. Keywords benefiting from search-lift attributable to the first content item are used to determine when to send a second content item in response to requests from client devices in a select audience. In some implementations, requests are compared to identify a set of keywords invoked more frequently after presentation of the first content item where an increase in usage exceeds a threshold or otherwise indicates a statistical significance.

Abstract: In one embodiment a method for probabilistically quantifying a degree of relevance between two or more citationally or contextually related data objects, such as patent documents, non-patent documents, web pages, personal and corporate contacts information, product information, consumer to behavior, technical or scientific information, address information, and the like is provided. In another embodiment a method for visualizing and displaying relevance between two or more citationally or contextually related data objects is provided. In another embodiment a search input/output interface that utilizes an iterative self-organizing mapping technique to automatically generate a visual map of relevant patents and/or other related documents desired to be explored, searched or analyzed is provided.

Abstract: Methods, systems, and computer program products for structured representation and classification of noisy and unstructured tickets are provided herein.

Abstract: Computer program products and systems are provided for mining for sub-patterns within a text data set. The embodiments facilitate finding a set of N frequently occurring sub-patterns within the data set, extracting the N sub-patterns from the data set, and clustering the extracted sub-patterns into K groups, where each extracted sub-pattern is placed within the same group with other extracted sub-patterns based upon a distance value D that determines a degree of similarity between the sub-pattern and every other sub-pattern within the same group.

Abstract: A method relates to receiving, by a processing device, a document comprising a first topic to be imported into a content management system, calculating a first signature of the first topic in view of content associated with the first topic, determining whether the first topic of the document is substantially similar to at least one of a plurality of topics stored in the content management system by comparing the first signature with a respective signature of the plurality of topics stored in the content management system, and in response to a determination that the first topic of the document is not substantially similar to any of the plurality of topics stored in the content management system, adding the first topic and the content associated with the first topic to the content management system.

Abstract: A method and a system for summarization of short comments are provided. The system comprises a memory to store a plurality of comments. Each of the plurality of comments including an overall rating and at least one phrase. The system also includes one or more processors to implement an aspect module to identify a head term based on a portion of the plurality of comments and to map the portion of the plurality of comments to an aspect corresponding to an attribute of an entity. The one or more processor also implement an extraction module to extract, from the portion of the plurality of comments, a representative phrase corresponding to the aspect.

Abstract: Architecture that splits the generation of results summaries into at least two phases for reduced latency and improved relevance. A first phase generates a summary quickly and thereby enables subsequent modules to begin processing. A second round then executes in parallel to these other modules to offset the latency. The second round can also provide additional contextual information to the summarization module to improve the quality (relevance) of the summaries. Ultimately, the summaries generated in the two phases are merged. The multiple rounds of summarization enable the first round to be cheaper in processing resources to save overall latency, whereas the second phase can be more expensive, since the second phase can be run in parallel with other modules. Additionally, contextual information can be used to build the full content summaries in the second round.

Abstract: A system for synonym generation and/or identification can be utilized to make a search engine more effective in finding relevant search results. In embodiments, a synonym generation system includes a phrase vector module, a vector similarity module, and a vector filter module. Candidate synonym phrase pairs are selected from data sources for analysis. Data sources may include a log of search queries, a corpus of web text, and a set of merchant descriptions of products. The data sources may be analyzed with respect to the phrase pairs to generate vector representations of the phrase pairs. The vector representations may then be analyzed to determine a similarity vector. The similarity vector allows the synonym generation system to filter synonyms from the candidate phrase pairs.

Abstract: Embodiments are described herein that provide searches, including a multi-dimensional search, a cross-source search, or both in a media playback system. The search can be initiated by way of a selection of a location on user interface of a controller. The location corresponds to one or more metadata that is used in the search. Results are sorted and displayed. In some embodiments, the one or more metadata is used to filter and/or sort the results.

Abstract: Disclosed herein is an automated approach for summarizing media content using descriptive information associated with the media content. For example and without limitation, the descriptive information may comprise a title associated with the media content. One or more segments of the media content may be identified to form a media content summary based on each segment's respective similarity to the descriptive information, which respective similarity may be determined using a media content and auxiliary data feature spaces. A shared dictionary of canonical patterns generated using the media content and auxiliary data feature spaces may be used in determining a media content segment's similarity to the descriptive information.

Abstract: A mobile terminal includes a display and controller configured to cause the display to display a playback screen of a first multimedia file and cause the display to display a first retrieval screen in response to receiving a first user input during the displaying of the playback screen of the first multimedia file, such that the first retrieval screen includes a plurality of thumbnail images respectively corresponding to one of a plurality of playback points in time on a per first time interval basis of the first multimedia file. The controller also causes the display to play the first multimedia file beginning at a playback point in time that corresponds to a selected one of the plurality of thumbnail images.

Abstract: Deeplinks relevant to the context of a search query are included with a search result returned in response to the search query. Deeplink hierarchies may be generated for websites by analyzing web pages within the websites. A deeplink hierarchy for a website includes a hierarchy of nodes with each node including a group of deeplinks to web pages within the website. When a search query is received and a web page result identified, the search query may be employed to identify a relevant node of deeplinks from a deeplink hierarchy for a website corresponding with the web page result. A search result may then be returned for the web page result that includes the identified deeplinks.

Abstract: In one exemplary embodiment, a computer-implemented method includes searching a database of annotations associated with an annotating proxy server. An annotated web page is obtained. The annotated web page can include an annotated web-page element. An annotation of the annotated web-page element can be included in the annotated web page by an annotating proxy server. The annotation can be stored in a database of web page annotations. The annotated web page can be annotated with a query field. A user-generated search query can be received in the query field. The database of web page annotations can be searched according to the user-generated search query.

Abstract: A centralized control center system for searching and managing one or more objects on granular object level in one or more datacenters is provided. The control center system includes a memory unit, and a processor. The memory unit stores a database and a set of modules. The processor executes the set of modules. The set of modules includes a search module, a topology view module, and an orphan object visibility module. The search module is configured to search and retrieve the one or more objects based on a search query. The topology view module is configured to display the one or more objects connected to the one or more datacenters in a hierarchy. The orphan object visibility module is configured to display (i) an object without a parent object, and (ii) a pool without a parent pool.