13 results about "Reference type" patented technology

The invention discloses a data processing method and device, a storage medium and electronic equipment, and the method comprises the steps: obtaining input multi-modal problem data in a question and answer scene, classifying the multi-modal problem data into text type problem data and reference type problem data through a large multi-modal data processing model, and carrying out the classification of the text type problem data and the reference type problem data; determining text type data features of the text type question data and reference type data features of the reference type question data, performing feature fusion processing on the text type data features and the reference type data features to obtain fused data features, and generating answer data for the multi-modal question data based on the fused data features, answer data for the multi-modal question data is displayed, and a reference modal class corresponding to the reference class question data is different from a text modal class of the text class question data. According to the method, the accuracy of the answer data generated for the multi-modal question data in the question and answer scene can be improved.

This disclosure proposes a text processing method, apparatus, electronic device, and storage medium, relating to the field of computer technology. The text processing method includes: acquiring a text to be processed; processing the text to be processed based on a reference processing pattern to obtain multiple reference subtexts, wherein the reference processing pattern has a corresponding processing type, and the type of the reference subtexts is the same as the processing type; adding at least one of the multiple reference subtexts to an initial substring to obtain a reference substring, wherein the reference substring has a corresponding reference scale value; determining the reference type of the last character of each added reference subtext; and determining the processing result corresponding to the text to be processed based on the reference scale value, reference type, and reference substring. This addresses the technical problems of inaccurate text processing and poor text processing results in existing technologies.

The application discloses a kind of data anomaly detection method, system, equipment and medium, by selecting the user data corresponding to each user in user service demand data from preset database.User service demand data is used to identify reference type, determine the reference type of judgment.If the reference type of judgment is first judgment reference, meter rotation data anomaly identification is carried out using user data, to generate first data anomaly detection result.If the reference type of judgment is second judgment reference, industry expansion suspension work order data anomaly identification is carried out using user data, to generate second data anomaly detection result.If the reference type of judgment is third judgment reference, basic file data anomaly identification is carried out using user data, to generate third data anomaly detection result.Based on the reference type of judgment, corresponding data anomaly identification is carried out using user data, quickly find out the error caused by manual operation process, improve the working accuracy of business, the working accuracy of business personnel.

The application discloses a target detection method and device, computer equipment and a computer readable storage medium, and belongs to the technical field of aircrafts. The method comprises the following steps: obtaining an existing tracking result; determining a target to be measured according to the existing tracking result; detecting a target belonging to a reference type in the target to be measured based on a reference characteristic possessed by the target of the reference type, the target of the reference type being a false target detected according to a reference false echo, and the reference false echo being an echo formed due to reflection of electromagnetic waves emitted by a radar by an aircraft. In this way, the target belonging to the reference type can be detected and timely suppressed, so that the target belonging to the reference type can be used to provide more reliable flight guidance for flight of the aircraft, and the aircraft can be prevented from mistakenly avoiding obstacles, thereby improving flight stability of the aircraft.

Techniques are disclosed for projecting information between runtime objects and memory regions. To this end, the system maps elements of a reference type to constituent layout objects of a compound layout object. Based on the mappings, the system determines a first method for extracting data from a source memory region described by the compound layout object and generating a target runtime object of the reference type to represent the extracted data in a runtime memory area that is managed by a garbage collector. Furthermore, based on the mappings, the system determines a second method for writing data represented by a source runtime object of the reference type to a target memory region in accordance with the compound layout object. The system subsequently uses the first method and / or the second method to project information between runtime objects of the reference type and memory regions corresponding to the compound layout object.

The invention relates to a live broadcast content dynamic director method and system based on an audience behavior graph, and relates to the technical field of the Internet, and the method comprises the steps: collecting a detection text and interaction information; generating a fluctuation curve based on the interaction information; identifying a text type from the detection text; combining the text type with a preset reference type to obtain a mark type; calling a reference fluctuation curve according to the mark type; comparing the consistency of the fluctuation curve and the reference fluctuation curve to obtain a future fluctuation curve, and calling maximum interaction information in combination with the interaction information; calling an output time point from the future fluctuation curve; obtaining an interaction picture according to the maximum interaction information and the fluctuation curve; and combining the interaction picture and the mark type to obtain an interaction instruction, switching the interaction picture, and outputting the interaction instruction at the output time point. The method has the effect of improving the accuracy of live broadcast interaction.

Timing information exchange for enhanced sidelink positioning is described. An apparatus is configured to identify a synchronization reference type of a synchronization reference source based on synchronization of the UE with the synchronization reference source. The apparatus is configured to provide a first indication of the synchronization reference type to at least one of the LMF or another UE, the quality metric or the time of change therefrom as a basis for communication. Another apparatus is configured to receive, from at least one of the LMF or another UE, at least one of a first indication of a synchronization reference type of a synchronization reference source associated with synchronization of the UE, a quality metric, or a time of change. The apparatus is configured to communicate based on the first indication of the synchronization reference type, the quality metric, or the change time.

The embodiment of the invention discloses a lightning electromagnetic wave waveform identification method and related equipment, and the method comprises the steps: calling a plurality of typical reference waveform sequences from a preset reference type library, calculating a similarity measurement value of a sampling sequence and each reference waveform sequence, and taking a difference value between a similarity maximum value and a similarity secondary maximum value as a judgment basis. Parameters such as waveform amplitude, pulse width and rising edge do not need to be used for judgment, misjudgment easily generated when different types of waveform feature intervals are overlapped is avoided, the waveform identification accuracy is improved, meanwhile, only a preset reference waveform library and similarity calculation logic are needed, complex model training and a large number of calculation resources are not needed, and the method is easy to implement. And rapid deployment in the low-computing-power online monitoring terminal is realized.

A mechanism for processing media data is disclosed. In one aspect, a method includes determining that a track identifier (ID) identifying a reference track of the media data is included in a track reference type box (TrackReferenceTypeBox) having a reference type (reference_type) equal to an associated external stream track ('aest'). The method also includes performing a conversion between the media data and a media data file based on the track identifier.

The application provides a method for recording object references based on threads in Java, comprising the following steps: obtaining all reference types of C++ holding Java objects in a system; creating reference interfaces according to the reference types respectively; classifying all reference interfaces to obtain a plurality of reference interface sets; creating encapsulation classes respectively to encapsulate any reference interface set; and establishing statistical interfaces in the encapsulation classes respectively to count the reference quantity corresponding to any reference type, wherein the application counts the reference quantity corresponding to different reference types according to all reference types of C++ holding Java objects, and then obtains the resource leakage situation of any reference type, which is beneficial to investigation according to the reference types.

The application discloses a data storage method and device, a computing device and a readable storage medium, and relates to the computer field.The data storage method comprises the following steps: detecting whether to-be-stored data is cross-process shared data; if the to-be-stored data is cross-process shared data, storing the to-be-stored data by using a shared storage class; if the to-be-stored data is non-cross-process shared data, obtaining the type of the to-be-stored data; when the to-be-stored data is a reference type or a large object data type, storing the to-be-stored data by using a file cache; and when the to-be-stored data is a basic data type and is not a large object data type, storing the to-be-stored data by using a lightweight storage class, wherein the large object data type is a data type whose memory occupation exceeds a preset value.The application supports the storage of multiple types of data, and can improve the data access efficiency.

The application provides a reference type analysis method and system based on joint multi-task learning, and relates to the technical field of semantic classification. The method comprises the following steps: configuring input data and label data of a reference intention classification task; defining a reference evaluation classification task, configuring input data and label data of the reference evaluation classification task; constructing a joint multi-task learning framework, embedding the reference intention classification task and the reference evaluation classification task as parallel classification tasks into the joint multi-task learning framework; adopting an alternating iteration strategy of the parallel classification tasks, training the joint multi-task learning framework, balancing gradient propagation between the parallel tasks, obtaining a reference type joint analysis model, and performing reference type analysis processing. Through the application, the technical problem that the deep semantic correlation between the reference intention and the reference evaluation cannot be effectively mined in the prior art, thereby affecting the intelligent analysis capability of scientific research content, can be solved, and the technical effect of improving the interpretation capability of the citation semantics can be achieved.

This application provides a file access method and related apparatus. In this method, in response to a first access request to access a first base file, a first file identifier and a first index number are obtained from the first file control parameters corresponding to the first base file. The first file identifier is the identifier of a first data file corresponding to the first base file. The first data file stores a first security attribute of the first base file. The first index number indicates the location of the first security attribute within the first data file. If the file type of the first data file determined by the first file identifier is an access rule reference type, the first security attribute in the first data file is read according to the first index number to access the first base file. In this application, when the type of the first data file is an access rule reference type, the first security attribute in the first data file is directly read according to the first index number, improving access efficiency.