Huawei Technologies Co., Limited

Sap Se

Fih (hong Kong) Limited

Chao Yu

Wei Dong Chen

Xiyu Zhou

Bifeng Tong

Chengliang Zheng

Qiwei Zhang

Ke Wang

Guangji Zhu

Chang Yi

Qiong Luo

Ling Chen

Hao Chen

Yinling Ni

Wei Fang

Ye Jin

Chen Gong

Jing Liu

Biao He

Chuanfei Xu

Zhengyuan Peng

Xinyu Zhou

A model selection method and a terminal are provided. The method includes: A terminal obtains first behavior data generated when a user operates the terminal. The terminal selects a first model from a plurality of stored models. The terminal runs the first model by using the first behavior data as an input parameter of the first model, to obtain a first output result. The terminal provides a first artificial intelligence AI service based on the first output result. When the first model does not meet a condition, the terminal selects a second model from the plurality of models. The terminal runs the second model by using second behavior data as an input parameter of the second model, to obtain a second output result. The terminal continues to provide the first AI service based on the second output result.

Model selection method and terminal

A method for identifying questions with same semantics includes creating a user behavior graph based on at least two user logs; determining a first associated question node set based on the user behavior graph and a seed question; calculating, based on the first associated question node set, association scores of questions corresponding to all question nodes in the user behavior graph; using a question corresponding to each associated question node in the first associated question node set as a new seed node, to determine M second associated question node sets; updating, based on the M second associated question node sets, the association scores of the questions corresponding to all the question nodes in the user behavior graph; and determining K questions with top K association scores based on updated association scores of the questions corresponding to all the question nodes in the user behavior graph.

Method for identifying questions with same semantics and electronic device

A system startup method includes creating a first thread when a kernel driver in a kernel mode detects a first disk partition, reading, in the kernel mode, metadata of the first disk partition using the first thread, and writing the metadata of the first disk partition into a first page cache using the first thread. In the kernel mode, metadata of a disk partition is pre-cached into a page cache using the first thread, and in a subsequent process in a user mode, the metadata of the disk partition is directly read from the page cache. A storage area of the page cache is memory and has a higher read/write operation rate compared with a disk.

System startup method and related device

A storage fragment management method and a terminal. The method may be applied to a file system of the terminal, and the file system includes at least one segment. The method includes: first determining, by the terminal, a source segment from the file system based on an aging degree of the segment and a valid block ratio of the segment; then determining, by the terminal from the file system based on an aging degree of the source segment, a target segment whose aging degree is consistent with the aging degree of the source segment; and finally migrating, by the terminal, data of a valid block in the source segment to an idle block in the target segment. This method is used to resolve a problem that power consumption is high when data migration is performed on a storage fragment in an existing log-structured file system.

Storage fragment management method and terminal

A garbage data scrubbing method includes obtaining an input/output (IO) busy/idle status of a terminal at a current moment, where the IO busy/idle status includes a busy state and an idle state. When the IO busy/idle status of the terminal at the current moment is the idle state, a discard message is delivered to a storage device, where the discard message includes an initial address and a size of to-be-scrubbed physical space in the storage device, and where the discard message is used to unbind a mapping relationship between a physical address of the to-be-scrubbed physical space and a corresponding logical address.

Garbage data scrubbing method, and device

A method for protecting application data includes detecting a deleting operation performed by a first application on first data. The method also includes determining, based on a preset condition, whether to delete the first data of the first application. The method further includes renaming the first data of the first application based on a determination that the first application meets the preset condition. The method additionally includes storing the renamed first data in a memory.

Method for protecting application data and terminal

This application provides a garbage data scrubbing method and a device, and relates to the field of terminals, to resolve a problem that delivering a discard message in a file system transaction affects a user foreground operation. The method includes: obtaining an IO busy/idle status of a terminal at a current moment, where the IO busy/idle status includes a busy state and an idle state (S); and if the IO busy/idle status of the terminal at the current moment is the idle state, delivering a discard message to a storage device (S), where the discard message includes an initial address and a size of to-be-scrubbed physical space in the storage device, and the discard message is used to unbind a mapping relationship between a physical address of the to-be-scrubbed physical space and a corresponding logical address.

Disclosed herein are system, method, and computer-readable device embodiments for automatically correcting erroneous license plate numbers generated by automatic license plate recognition. An embodiment operates by selecting a set of adjacent license plate recognition (LPR) stations comprising a first LPR station, a second LPR station, and a third LPR station, accessing an error pattern probability matrix for at least the third recognized license plate number accessed in relation to at least one of the first recognized license plate number or the second recognized license plate number differing from the third recognized license plate number, and determining a corrected license plate number based on the error pattern probability matrix. In some embodiments, LPR correction may be accurately realized even when an erroneous license plate numbers are missing characters, such as by visual obstruction from an LPR camera or sensor, or has no characters in common with the real license plate number.

Automatic detection and correction of license plate misidentification

Disclosed herein are system, method, and computer program product embodiments for automatically detecting cloned vehicle identifiers. An embodiment operates by determining legitimate trajectory information and illegitimate trajectory information corresponding to vehicular travel in a geographic area, and training a decision model to detect a cloned vehicle identifier based on the legitimate trajectory information and the illegitimate trajectory information. Furthermore, the embodiment identifies vehicles potentially using a cloned vehicle identifier based on the decision model and common driving patterns within the geographic area. In some embodiments, enforcement agents are notified of the vehicles potentially using a cloned vehicle identifier.

Automatic identification of cloned vehicle identifiers

Described herein is a descriptive framework to facilitate error detection in recognition data. In accordance with one aspect of the framework, at least one erroneous record is detected in a first set of recognition data. The framework may determine a correction of a first recognized identifier in the erroneous record by searching a second set of recognition data for a matching record with a second recognized identifier substantially similar to the first recognized identifier. A report may then be generated to present the detected erroneous record and the determined correction.

Growing community of inventors

Nanjing, China

Chao Yu