An information processing method includes obtaining text information and a sentence set; encoding a sentence in the sentence set using a first encoder to obtain a first encoded vector, and encoding the sentence using a second encoder to obtain a second encoded vector. The first encoded vector is determined according to the sentence, and the second encoded vector is determined according to a feature of the sentence. The method also includes determining a sentence encoded vector according to the first and second encoded vectors; encoding the sentence encoded vector using a third encoder to obtain global information; decoding the global information using a decoder; and determining a probability value corresponding to the sentence. Accordingly, when a deep learning method is used, a manually extracted sentence is further added to perform feature training, to effectively improve a learning capability of a model, thereby improving an information processing capability and effect.

A computer-implemented method according to one embodiment includes dividing data into a plurality of partitions, creating a plurality of groups of different data types within each of the plurality of partitions, independently compressing, within each of the plurality of partitions, each of the plurality of groups of different data types to create a plurality of independently compressed partitions, and validating each of the plurality of independently compressed partitions to create a plurality of validated independently compressed partitions.

A system and method for a non-transient computer readable medium containing program instructions for causing a computer to perform a method for compressing data comprising the steps of receiving a data string for compression, the data string including a plurality of data elements, creating a template based on processing the data string, the template including common information across all data elements of the data string, creating one or more entries, wherein the one or more entries include information that is different to the template, and storing the template and the one or more entries.

A computer architecture for graph-traversal provides a processor for bottom-up sequencing through the graph data according to vertex degree. This ordered sequencing reduces redundant edge checks. In one embodiment, vertex adjacency data describing the graph may be allocated among different memory structures in the memory hierarchy to provide faster access to vertex data associated with vertices of higher degree reducing data access time. The adjacency data also may be coded to provide higher compression in memory of vertex data having high vertex degree.

A computer-implemented method according to one embodiment includes dividing data into a plurality of partitions, creating a plurality of groups of different data types within each of the plurality of partitions, independently compressing, within each of the plurality of partitions, each of the plurality of groups of different data types to create a plurality of independently compressed partitions, and validating each of the plurality of independently compressed partitions to create a plurality of validated independently compressed partitions.

Systems, methods, and media for managing an in-memory NoSQL database are provided. In some embodiments, systems for managing an in-memory NoSQL database are provided, the systems comprising a hardware processor that is configured to: identify a group of keys capable of being included in a complex multi-key database command, the group including a first key and a second key, the first key being associated with a first value and the second key being associated with a second value, the first value and the second value being stored on different servers that are part of the database; and cause the first value and the second value to be both stored on a given server.

Unsorted sparse dictionary encodings are transformed into unsorted-dense or sorted-dense dictionary encodings. Sparse domain codes have large gaps between codes that are adjacent in order. Unlike spare codes, dense codes have smaller gaps between adjacent codes; consecutive codes are dense codes that have no gaps between adjacent codes. The techniques described herein are relational approaches that may be used to generate sparse composite codes and sorted codes.

Some embodiments provide a non-transitory machine-readable medium that stores a program. The program determines a scale value based on a plurality of floating point values. The program further scales the plurality of floating point values based on the scale value. The program also converts the plurality of floating point values to a plurality of integer values. The program further determines an integer encoding scheme from a plurality of integer encoding schemes. The program also encodes the plurality of integer values based on the determined integer encoding scheme.

Techniques are described herein for space-efficient encoding of label information of property graphs. In an embodiment, an input graph is received. The input graph comprises a plurality of entities and a plurality of label sets. Each entity of said plurality of entities is associated with a label set of the plurality of label sets and each label set of the plurality of label sets comprises zero or more labels of a plurality of labels. A first mapping is generated that maps each label of the plurality of labels to a label code. A second mapping is generated that maps each label integer set of a plurality of label integer sets to a label code. Each label integer set of the plurality of label integer sets corresponds to a label set of the plurality of label sets, wherein each label integer set of the plurality of label integer sets comprises label codes from the first mapping that are mapped to each label included in the corresponding label set. A compressed label set is generated for each entity of the plurality of entities. Each compressed label set comprises a plurality of bits that indicate a zeroth state, a first state, a second state, or a third state. The compressed label sets and the first and second mappings are used to efficiently evaluate graph label queries.

A method of encoding print data performed by a host device is described in which a print request for a compound document is received; objects included in the compound document is classified into predetermined groups based on object attribute information; each of the predetermined groups is compressed according to a preset compression method; and the compressed groups is merged and the merged groups is transmitted to an image forming apparatus.