Abstract: A method of generating a caption for a social media post is provided. The method may include receiving a social media post to be posted to a social media network; collecting reference data relevant to determining common activities occurring at a category of location associated with the social media post; inferring potential topics for captions from a topic inference model, based on the collected reference data associated with the social media post; setting parameters associated with a language model based on the inferred topic; and generating at least one caption for the social media post based on the identified language model, and the inferred topic.

Abstract: Described are a method, system, and computer program product for deduplicating a storage object. A hash of a window of data of a storage object is determined and a determination is made as to whether the window of data of the storage object corresponds to a chunk boundary. A determination is made as to whether the hash of the object matches one pseudo fingerprints in a list of at least one pseudo fingerprint. A storage object chunk boundary based on the window of data is stored in response to the window of data corresponding to the chunk boundary or in response to determining that the hash of the object matches one of the pseudo fingerprints. A determination is made of a new window of data in the storage object following the window of data when the window of data is not an end of data of the storage object.

Abstract: A content recommendation system includes: an attribute value storage means which stores one or more attribute values of attributes for each content; a preference distribution storage means which stores a user preference degree for each attribute value; an attribute value acquisition means which acquires an attribute value of at least one attribute for each content; a preference distribution update means which updates the storage contents in the preference distribution storage means in accordance with the acquired attribute value; a condition determination means which selects an attribute value of one or more attributes in accordance with probability based on the user preference degree stored in the preference distribution storage means and determines the condition of the attribute value of the one or more attributes in accordance with the selected attribute value; and a content selection means which selects some or all of the contents in accordance with the determined condition.

Abstract: A system and method for performing an image level restore of data is described. In some examples, the system receives a request to restore a file and transmits the request to an intermediate component. The intermediate component may then retrieve a directory file from an image of a secondary copy of a data set, identify a location of the file from the directory file, and provide the location to the requestor.

Abstract: The present invention provides a method and apparatus for searching information. The method is for, in response to receipt of a keyword from a search user of a plurality of users, searching in an information repository to obtain one or more information items to be returned; and returning the information item to the search user based on an importance of each information item of the one or more information items. With the method, it is possible to fully leverage various objective internal relationships between search users, information items and creators and thus improve the search accuracy.

Abstract: Aspects include creating a profile for a user from user account data and a user activity log, searching an indexed repository for keywords matching elements from the user account data and the user activity log, and retrieving documents from the indexed repository that are identified based on the searching. Aspects further include creating a custom portal for the user that includes the documents.

Abstract: An approach is provided for integrating place metadata provided by a community of metadata builders, including receiving registration data that indicates one or more values for a corresponding one or more attributes that describe a place. A place is associated with a geographic location. Providing an indication of match between the registration data and metadata for a predetermined place is also initiated. In some embodiments, a new entry for a set of predetermined places is generated based on validating the registration data and a negligible degree of match. In some embodiments, a unique identifier for the place is included in indication of match for either a new place represented by the registration data or a matching predetermined place.

Abstract: Multi-tenant access to respective isolated data sets organized using different application schemas is disclosed. Multiple tenants store respective data sets in separate datastores. The datastores are organized using different application schemas, each derived from a same base schema. However, the tenants share access to a same application. The application receives a first request from a first tenant and a second request from a second tenant. The application serves the first request by accessing a first data set organized using a first application schema. The application serves the second request by accessing a second data set using a second application schema.

Abstract: In the context of tracking systems, it is difficult to ensure that an organization has a complete, accurate database of contacts stored in its tracking system. When tracking systems users are required to manage exporting and importing of contacts from their desktop mail clients and handheld devices, it is almost certain that contact information will not be kept up-to-date and that confidence in the accuracy of the contact information will not be high. By enabling a remote directory access portal in the tracking system, all users can be assured that they have available the latest contact information for the organizations' contacts. In addition to providing directory access, the tracking system can authenticate users and, based on the users' entitlements, authorize users' access to specific contacts.

Abstract: Techniques are disclosed for accessing on-premise and off-premise datastores organized using different application schemas. An interface serves requests, from a tenant, by accessing either the on-premise datastore or the off-premise datastore. The interface receives a request for a particular data set. Based on the request itself, the interface selects either the on-premise datastore or the off-premise datastore to retrieve the particular data set. The interface retrieves the particular data set from the selected datastore without attempting to retrieve the particular data set from the non-selected datastore. The interface transmits and/or presents the particular data set.

Abstract: A system that implements a data storage service may store data on behalf of storage service clients. The system may maintain data in multiple replicas that are stored on respective computing nodes in the system. Updates to the stored data and to the membership of replica groups are propagated as replicated log records. A replica receiving a log record may compare metadata in the received log record to corresponding metadata in a log record that was previously appended to its log to determine a response. The metadata may include a sequence number, a lock generation identifier, an epoch identifier, or an indication of an epoch change. The replica may append the received log record to its log, drop the received log record, or cache the received log record for future use. If a log conflict indicates an invalid log stream branch, one or more log records may be deleted.

Abstract: There are disclosed a system and method for providing a word search service based on message according to the present invention. The system for providing a word search service based on message according to the present invention includes a database which is equipped with a plurality of data modules and registers sentence data and word extracted from the sentence data in each data module equipped to manage the sentence data and the word; and a service server which, if receiving a search word from a user terminal, broadcasts the received search word to all data modules and is provided with a search result from each of all data modules in response to broadcasting and thereafter provides the provided search result to the user terminal.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying languages that are relevant to resources. In one aspect, a method includes selecting in a data processing apparatus a first resource; accessing click data that identifies, for each of a plurality of requests for the first resource, a respective search engine user interface from which the request was received; identifying a search engine user interface language for each of the plurality of requests based on the click data; determining a respective language relevance score for the first resource for each identified search engine user interface language; and selecting one or more languages as being relevant to the first resource based on the language relevance scores.

Abstract: A method and system for managing a search is disclosed. The method and system can include receiving a set of first results from a first query. A set of categories can be identified based on the set of first results. The method and system can include generating a set of second results using a second query having at least a portion of the first query, at least a portion of the first results, and at least a portion of the set of categories. The method and system can also include scoring the set of second results based upon a relationship of the first query, the second query, and the set of second results.

Abstract: A method and system for servicing a data query in a relational database management system (RDBMS). Access path security corresponding to a user originating the data query is configured. The data query requests data from the RDBMS pursuant to a received configuration query associated with the user. The data query requests a value stored in a result field by specifying the result field and a result table of a database in the RDBMS. The configuration query specifies a usable index field associated with the result table. The access path security mandates the user to provide a search key in the data query in searching the result table. The search key includes a name identical to the usable index field specified in the configuration query such that the access path security enables users to search the result table only when the users specify the usable index field in the data query.

Abstract: Method of retrieving event information is presented. Memento objects can be recognized by an archive engine. Based on the recognition, the archive engine obtains information related to the memento object, possibly one or more recognizable features, and uses the information to search for events associated with a timeline that have corresponding tags. The archive engine can then return the event information as a result set to a user.

Abstract: Based on a probability of summary metadata associated with an entity of a media hosting service being inaccurate, a determination is made as to whether to verify the accuracy of the summary metadata. In response to determining to verify the accuracy of the summary metadata, a determination is made as to whether the summary metadata is inaccurate with respect to master metadata associated with the entity of the media hosting service. The summary metadata is updated with a correct value determined from the master metadata, in response to determining the summary metadata is inaccurate.

Abstract: Tracking lineage of data. A method may be practiced in a network computing environment including a plurality of interconnected systems where data is shared between the systems. A method includes accessing a dataset. The dataset is associated with lineage metadata. The lineage metadata includes data indicating the original source of the data, one or more intermediary entities that have performed operations on the dataset, and the nature of operations performed on the dataset. A first entity performs an operation on the dataset. As a result of performing a first operation on the dataset, the method includes updating the lineage metadata to indicate that the first entity performed the operation on the dataset. The method further includes providing functionality for determining if the lineage metadata has been compromised in that the lineage metadata has been at least one of removed from association with the dataset, is corrupted, or is incomplete.

Abstract: An object logical node achieved by a distributed data management device includes a storage unit for storing object node identifiers each uniquely assigned to a plurality of logical nodes, a storage unit for storing at least one of a plurality of partial data units, a link table for storing link information between the object logical node and a link destination logical node, a storage unit for storing a value range boundary value with respect to each attribute corresponding to the partial data unit, and a storage unit for storing a tree structure data unit with respect to each attribute including a plurality of tree nodes each indicating a value range for identifying the logical node storing the partial data unit corresponding to an access request, the tree structure data unit including a root tree node including at least one entry formed with a pointer pointing to a child tree node associated with the link destination logical node and a value indicating a value range for selecting the pointer.

Abstract: A set of search requests may be analyzed to detect fixed phrases suitable for inclusion in a search index. Sets of candidate phrases may be identified among the search requests. Fixed phrases may be detected among the candidate phrases using statistical techniques, for example, by identifying phrases having a relatively high pointwise mutual information (PMI) with respect to component keywords. Fixed phrase detection may include keyword and/or phrase clustering. Clusters may correspond to topics defined using a latent Dirichlet allocation (LDA) procedure. Fixed phrase detection may include identifying phrases having relatively high PMI within particular clusters.