Abstract: A data processing method and apparatus are described. The method includes receiving a request which carries a user identifier (ID) from a user terminal. The method also includes determining all associated user IDs associated with the user ID. The method also includes selecting N particular user IDs which meet a set condition from all the associated user IDs, pulling latest user data corresponding to the N particular user IDs, and returning the latest user data corresponding to the N particular user IDs to the user terminal. Wherein N is smaller than or equal to a user data amount M requested by a page.

Abstract: Embodiments of the present invention provide systems and methods for detecting data redundancy within a database and optimizing data access operations. The embodiments identify a candidate column to determine the relationship between the candidate column against the remaining columns. The system calculates the vector angles of rows based on candidate column pairs and determines the difference in the angle between the candidate column and the corresponding rows of data in the comparison columns. If the candidate column has a greater angle than the compared column, then the compared column is identified as redundant and is marked for the decluttering process.

Abstract: A fusion condition providing system for an optical fiber fusion splicer includes: a storage device storing a database of fusion conditions; a user-side communication unit transmitting a command for requesting a fusion condition for a pair of optical fibers to be fusion spliced by a user's fusion splicer; and a data management unit. The data management unit, in accordance with the command received from the user-side communication unit via a network, transmits the requested fusion condition among the plurality of fusion conditions inside the storage device to the user-side communication unit via the network so that the requested fusion condition is available for use in the fusion splicer. If the requested fusion condition was not in the database, the data management unit updates the database by causing the requested fusion condition to be newly created and storing the newly created fusion condition in the database.

Abstract: A network communication stack running on relational processing circuitry performs control and maintenance actions on records from a database server managed by repository control circuitry. The database interaction layer of the communication stack accesses the records on the database server. The database interaction layer passes the accessed records to the data processing layer for parsing and storage as tabular entries. An operator may perform manipulations on the tabular entries using a command interface generated by the command layer of the communication stack using locally-defined interface parameters that are independent of characteristics of the database server. The data processing layer recompiles manipulated entries into an altered record. The database interaction layer sends the altered record back to the database server.

Abstract: Present-day blockchain architectures suffer from several problems, including poor extensibility and scalability. This may stem from tying two parts of the consensus architecture, namely canonicality and validity, too closely together. The Polkadot architecture, which is a heterogeneous multi-chain, provides better extensibility and scalability by setting canonicality and validity apart. Compartmentalizing canonicality and validity and keeping overall functionality to a minimum of security and transport introduces practical core extensibility in situ. Scalability is addressed through a divide-and-conquer approach to canonicality and validity, scaling out of its bonded core through the incentivization of untrusted public nodes. The heterogeneity of this architecture enables many highly divergent types of consensus systems interoperating in a trustless, fully decentralized “federation,” allowing open and closed networks to have trust-free access to each other.

Abstract: Kinship between electronic files among personal networked devices may be ascertained between the files by determining an operational relationship between the files and with a similarity measurement.

Abstract: A service recommendation method includes, when a user of a terminal requests a first service from an intelligent assistant, selecting, according to a name of the first service and by using a pre-established service relationship model, a potential service with a degree of relevance to the first service that meets a preset condition from multiple services that the intelligent assistant can provide, where names of the multiple services and degrees of relevance of the multiple services to each other are recorded in the service relationship model; and recommending the potential service to the user.

Abstract: A storage system communicatively coupled to a database management system (DBMS performs storage-side scanning of data sources that are not stored in native database storage format of the DBMS. Data sources for external tables are accessible in a storage system referred to as a distributed data access system (DDAS), e.g. a Hadoop Distributed File System. To execute a query that references an external table, a DBMS first generates an execution plan. The DDAS supplies the DBMS with information that specifies each portion of the data source, and specifies which data node to use to access the portion. The DBMS sends a request for each portion to the respective data node, requesting that the data node generate rows from data in the portion. The request may specify scanning criteria, specifying one or more columns to project and/or filter on, and code modules for the data node to execute to generate records.

Abstract: In an approach to content selection, one or more computer processors receive a first query from a user. The one or more computer processors retrieve a first response for the first query. The one or more computer processors determine whether the first response exceeds a threshold for a policy for content usage associated with the first response, wherein the policy for content usage specifies one or more restrictions for usage of a content asset. In response to determining the first response exceeds the threshold for the policy for content usage, the one or more computer processors select a first subset of the first response to meet the threshold for the policy for content usage. The one or more computer processors send the first subset of the first response to the user.

Abstract: Various aspects of the subject technology relate to systems, methods, and machine-readable media for language translation based on image search similarities. These aspects include an image retrieval system using a convolutional neural network that is trained to identify a correlation between an image and a language term, and using an image search engine to search against images corresponding to visual words that are responsive to a given search query in a given spoken language. These aspects include access to interaction probability data that identifies user interaction probabilities for the visual words to determine a correlation between the input language terms of the search query and the rate at which users interact with images of a corresponding visual word that is responsive to the search query. The system then provides a prioritized listing of images that is responsive to the given search query based on the identified user interaction probabilities.

Abstract: Techniques and systems are described to control a determination of document similarity. In one example, dimensionality of the documents is reduced through computation of a signature, e.g., via a hashing technique such as “minhashing” which is also known as min-wise independent permutations locality sensitive hashing. From these signatures, another hashing technique (e.g., locality sensitive hashing) is used to determine similarity of the signatures to each other. Identification of disjoint sets is then used as a basis to partition the documents for determination of document similarity by respective nodes of a plurality of computing devices. In this way, an amount of data shuffling between the nodes as part of the determination of document similarity may be reduced. In another example, a weighting is applied to attributes of documents as part of the determination of document similarity.

Abstract: In an embodiment, a system can determine geocoded data from a database of geographic coordinates and metadata. The system correlates metadata, such as invoice data, to geolocation data, such as GPS or cellular data to determine geocoded data. The system further identifies one or more geographic coordinates for one or more location names, which may not have a corresponding metadata entry, by generating clusters of geographic coordinates. The clusters are then matched to one or location names using a matching algorithm. Accordingly, improved geocoded data may be determined.

Abstract: Embodiments are described for a method and system for improving B+Tree scan performance by receiving a data access instruction that specifies pages to be accessed in a data store utilizing a B+Tree data structure; defining a read-ahead context comprising an array of page numbers corresponding to the specified pages; loading the read-ahead context array into a read-ahead cache; and reading the first page of the read-ahead context in a synchronous manner, and each of the subsequent pages of the read-ahead context in an asynchronous manner.

Abstract: A method trains a model for providing content items to users of a social networking system. The system generates profile vectors based on user profile information such as demographic data and personal data. The system logs actions performed by users on the social networking system and generates behavior vectors based on the logged actions. The profile vectors and behavior vectors are each associated with a user attribute, e.g., the age or gender of a user. The system generates a difference vector based on a profile vector and a behavior vector. The difference vector is then used as a feature to train the model using machine learning techniques. The trained model may select content items that a target user is most likely to be interested in and interact with.

Abstract: Systems, apparatuses, and methods for cloud memory sizing are disclosed. An initial database memory allocation is determined for the provisioning of a database server instance. Periodically, sizes of key database tables of the database server instance are measured and an upper and a lower bound ratio are determined based on the key database table sizes and a buffer pool size. The upper and lower bound ratios are used to determine a desired memory allocation from which a report is generated including an interface for generating an instance move action for re-provisioning the database server instance with the desired memory allocation.

Abstract: A method for recommending keywords can receive a first search term entered by a user, search a keyword library comprising a plurality of keywords and retrieve a preset number of keywords based on a similarity coefficient between each keyword and the first search term. After receiving a second search term entered by the user, the method obtains a correlation value between the second search term and the first search term based on whether a webpage in a search result of the first search term visited by the user includes the second search term, and determines the similarity coefficient between the second search term and the first search term in accordance with the correlation value. And then, the method updates the keyword library to save the similarity coefficient between the second search term and the first search term.

Abstract: Various embodiments provide a method and an apparatus for identifying webpage type. The method includes: judging whether a web address to be classified matches with a webpage classification rule in at least two webpage classification rules; and determining the type of the webpage to be a type corresponding to a webpage classification rule which matches with the web address.

Abstract: A method and a storage device for collecting garbage data, where the method includes separately recording a data volume of first data in each segment of a storage device at a current time and a variation of the first data in each segment of the storage device in a preset period of time before the current time; obtaining, by means of calculation, a predicted value of the first data according to the data volume of the first data in each segment and the variation of the first data in each segment; and determining, according to the predicted value of the first data in each segment, a segment whose garbage data needs to be collected. A segment that has more garbage data and whose garbage creating rate is lower is reclaimed preferentially.

Abstract: Methods and systems for searching enterprise data are provided. The method is for searching data and includes identifying edges associating data and/or people. The method also includes assigning a resistance value to an edge based on strength(s) of association, the resistance value being inversely related to the strength of the association. The method further includes forming a resistance graph connecting the user with the data and the people based on the values of the edges. Additionally, the method includes providing results for a search query from a user based on the resistance graph with lower resistance values being prioritized for the user.

Abstract: According to an example, at least one hot account is determined for each category according to quality scores and correlation degrees of history user generated content (UGCs); after a UGC newly posted by the hot account is received, if a quality score of the newly posted UGC is higher than a predefined quality score threshold and a correlation degree between the newly posted UGC and the category that the hot account belongs to is higher than a predefined correlation degree threshold, the newly posted UGC is determined as a hot UGC.