CN110750992A

CN110750992A - Named entity recognition method, device, electronic equipment and medium

Info

Publication number: CN110750992A
Application number: CN201910952910.2A
Authority: CN
Inventors: 杨志伟; 朴海音; 詹光; 陈贺昌; 常毅
Original assignee: Jilin University
Current assignee: Jilin University
Priority date: 2019-10-09
Filing date: 2019-10-09
Publication date: 2020-02-04
Anticipated expiration: 2039-10-09
Also published as: CN110750992B

Abstract

The application discloses a named entity identification method, a named entity identification device, electronic equipment and a medium. After obtaining data to be recognized, converting the data to be recognized into word embedding vectors based on a pre-trained word embedding vector matrix; performing multi-level feature extraction on the word embedding vector by using a plurality of neural network models and an attention mechanism to obtain feature data to be identified in different levels; and carrying out sequence marking on the characteristic data to be identified to obtain an entity identification result. By applying the technical scheme, compared with the prior art, the method and the device have the advantages that breadth learning and attention mechanism are combined to carry out named entity recognition, and then the effect of improving recognition accuracy can be achieved.

Description

Named entity recognition method, device, electronic equipment and medium

Technical Field

The present application relates to data processing technologies, and in particular, to a named entity identification method, apparatus, electronic device, and medium.

Background

Knowledge is the result of human recognition, which is embedded in the data. In the internet era, human beings generate large-scale data in the interaction process with the nature and the society, and how to mine valuable knowledge from the data and provide accurate knowledge service for users is a key difficulty for constructing an intelligent information service system. With the rapid development of artificial intelligence, Knowledge Graph (Knowledge Graph) formally proposed by Google in 2012 greatly promotes the development of semantic web, natural language processing and the like, becomes the basis of Knowledge-driven intelligent application, and is beneficial to providing more accurate and efficient information service.

With the development of natural language processing techniques, named entity recognition can be used to extract some kind of entity from a large amount of literature. At present, a named entity recognition method mainly comprises four strategies, namely dictionary-based, rule-based, statistical model-based, deep learning-based and the like. The recognition method based on the dictionary and the rule usually needs to manually formulate the dictionary and the rule, and extracts the entity by summarizing the rule of the entity and the context rule thereof, but because the entity recognition process has no standard word bank and needs to be built by self, the method is difficult to realize automatic recognition and has poor effect. Recognition methods based on statistical models include machine learning models such as HMM, MEMM and CRF, which are suitable for recognizing repeated entities and are often used for extracting phrase structures such as names of people and places from texts, but the method is not completely suitable for all entity recognition tasks.

Disclosure of Invention

The embodiment of the invention provides a named entity identification method, a named entity identification device, electronic equipment and a medium.

According to an aspect of the embodiments of the present application, there is provided a named entity identifying method, including:

acquiring data to be identified;

converting the data to be recognized into word embedding vectors based on a pre-trained word embedding vector matrix;

performing multi-level feature extraction on the word embedding vector by using a convolutional neural network, a recursive neural network and an attention mechanism to obtain feature data to be identified in different levels;

and carrying out sequence marking on the characteristic data to be identified to obtain an entity identification result.

Optionally, in another embodiment based on the foregoing method of the present application, after performing multi-level feature extraction on the word embedding vector by using a convolutional neural network, a recurrent neural network, and an attention mechanism to obtain different levels of feature data to be identified, the method further includes:

determining a weight parameter of each feature data to be identified based on the adjustment parameter;

and performing breadth feature fusion on each feature data to be identified based on the weight parameters of the feature data to be identified.

Optionally, in another embodiment based on the foregoing method of the present application, the performing sequence tagging on the feature data to be identified to obtain an entity identification result includes:

labeling the characteristic data to be identified with a label;

classifying each datum in the feature data to be identified based on a BIOES labeling mode;

and carrying out sequence labeling on the characteristic data to be recognized based on the label labeling and the category in the characteristic data to be recognized to obtain the entity recognition result.

Optionally, in another embodiment based on the foregoing method of the present application, the different levels of feature data to be identified include at least any one of:

global character features, global word features, local character features, local word features.

Optionally, in another embodiment based on the foregoing method of the present application, before the performing, by using a convolutional neural network, a recurrent neural network, and an attention mechanism, multi-level feature extraction on the word embedding vector to obtain different levels of feature data to be identified, the method further includes:

obtaining sample data, wherein the sample data comprises at least one entity naming feature;

and training a preset convolution neural network model by using the sample data to obtain the neural network meeting the fusion breadth attention of the preset condition.

According to another aspect of the embodiments of the present application, there is provided a named entity identifying apparatus, including:

the acquisition module is configured to acquire data to be identified;

a conversion module configured to convert the data to be recognized into word-embedded vectors based on a pre-trained word-embedded vector matrix;

the extraction module is configured to perform multi-level feature extraction on the word embedding vector by using a convolutional neural network, a recurrent neural network and an attention mechanism to obtain feature data to be identified in different levels;

and the generating module is configured to perform sequence marking on the characteristic data to be identified to obtain an entity identification result.

According to another aspect of the embodiments of the present application, there is provided an electronic device including:

a memory for storing executable instructions; and the number of the first and second groups,

a display for display with the memory for executing the executable instructions to perform the operations of any of the named entity recognition methods described above.

According to yet another aspect of the embodiments of the present application, there is provided a computer-readable storage medium for storing computer-readable instructions, which when executed perform the operations of any one of the named entity recognition methods described above.

In the application, after the data to be recognized is obtained, the data to be recognized can be converted into word embedding vectors based on a pre-trained word embedding vector matrix; performing multi-level feature extraction on the word embedding vector by using a plurality of neural network models and an attention mechanism to obtain feature data to be identified in different levels; and carrying out sequence marking on the characteristic data to be identified to obtain an entity identification result. By applying the technical scheme, compared with the prior art, the method and the device have the advantages that breadth learning and attention mechanism are combined to carry out named entity recognition, and then the effect of improving recognition accuracy can be achieved.

The technical solution of the present application is further described in detail by the accompanying drawings and examples.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description, serve to explain the principles of the application.

The present application may be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:

FIG. 1 is a schematic diagram of a system architecture of the named entity recognition method of the present application;

FIG. 2 is a schematic diagram of the named entity recognition apparatus of the present application;

fig. 3 is a schematic view of an electronic device according to the present disclosure.

Detailed Description

Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that: the relative arrangement of the components and steps, the numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present application unless specifically stated otherwise.

Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description.

The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the application, its application, or uses.

Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.

It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.

It should be noted that all the directional indications (such as up, down, left, right, front, and rear … …) in the embodiment of the present application are only used to explain the relative position relationship between the components, the movement situation, and the like in a specific posture (as shown in the drawing), and if the specific posture is changed, the directional indication is changed accordingly.

In addition, descriptions in this application as to "first", "second", etc. are for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicit to the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.

In this application, unless expressly stated or limited otherwise, the terms "connected," "secured," and the like are to be construed broadly, and for example, "secured" may be a fixed connection, a removable connection, or an integral part; can be mechanically or electrically connected; they may be directly connected or indirectly connected through intervening media, or they may be connected internally or in any other suitable relationship, unless expressly stated otherwise. The specific meaning of the above terms in the present application can be understood by those of ordinary skill in the art as appropriate.

In addition, technical solutions between the various embodiments of the present application may be combined with each other, but it must be based on the realization of the technical solutions by a person skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination of technical solutions should be considered to be absent and not within the protection scope of the present application.

A method for conducting named entity recognition according to an exemplary embodiment of the present application is described below in conjunction with fig. 1-2. It should be noted that the following application scenarios are merely illustrated for the convenience of understanding the spirit and principles of the present application, and the embodiments of the present application are not limited in this respect. Rather, embodiments of the present application may be applied to any scenario where applicable.

The application also provides a named entity identification method, a named entity identification device, a target terminal and a medium.

Fig. 1 shows a schematic flow chart of a named entity identification method according to an embodiment of the present application.

As shown in fig. 1, the method includes:

and S101, acquiring data to be identified.

First, it should be noted that, in the present application, a device for acquiring data to be identified is not specifically limited, for example, in the present application, the data to be identified may be acquired by an intelligent device, or may be acquired by a server. In addition, in the present application, the smart device is not specifically limited, that is, the smart device may be any smart device, such as a mobile phone, an electronic notebook, a PDA, and the like.

In addition, the data to be recognized is not specifically limited, that is, the data to be recognized can be any natural language text. The method and the device can judge the naming recognition result based on the data to be recognized. Specifically, named entity recognition is also called named recognition, is a basic task in natural language processing, and has a wide application range. Named entity recognition tasks are the recognition of entities that appear in sentences, where a named entity generally refers to an entity in text that has a particular meaning, such as a person's name, place name, organization name, etc.

S102, converting the data to be recognized into word embedding vectors based on the pre-trained word embedding vector matrix.

Further, one solution to convert an input sequence into a vector is by one-hot encoding, which is the most primitive way to represent characters and words. For simplicity, we take words as an example, and characters are similar. For example, the vocabulary is five words of "Bill, lives, in, New, York", and one-hot is to encode the five words by 0-1, as shown in table 1:

obviously, when a long input sequence is processed, the traditional one-hot coding is too sparse, and the resources are excessively occupied. Meanwhile, because the dimensions are independent of each other, semantic information cannot be obtained.

In order to overcome the semantic exact problem of one-hot coding, an embedded representation layer is introduced, and an original input space is mapped to another space with smaller dimension, so that not only can the semantic features of words be obtained, the words are no longer isolated individuals, but also the data volume required by training can be reduced. Thus, here semantic and syntactic features are better captured by randomly initializing character embedding and pre-training word embedding. However, the embedded representation is not limited to these two ways, and combining more different embedded representation ways can provide more feature sources, and theoretically, better entity recognition effect can be obtained.

It should be further noted that other pre-trained word vectors may also be utilized in the present application to establish a mapping dictionary from words to word vector numbers, and to map each word in the text to a corresponding word number. And establishing a word vector matrix, wherein the row number of each row corresponds to a corresponding word number, and each row represents a word vector. Assuming that there are N words in chinese, the word vector matrix can be represented as a matrix of N x d, where d represents the dimension of the word vector, and each word can be represented by a d-dimensional vector, i.e., ei. And splicing the word vectors to obtain sentence representation. For the input text, assuming that there are n words in the sentence, each word is represented by a d-dimensional vector, and the word vectors of all the words in the sentence are spliced to obtain an input matrix of the encoder, where the input matrix may be represented as x.

Before the advent of word-embedded vector technology, convolutional neural networks also used a one-hot encoded form of words as input, but one-hot encoding only represented the position of words in the vocabulary, had no practical meaning, and most importantly, could not represent word-to-word associations. The word embedding vector maps each word to a point in a high-dimensional space, different words are located at different positions, and words with close meanings are located closer in position.

And S103, performing multi-level feature extraction on the word embedding vector by using a convolutional neural network, a recurrent neural network and an attention mechanism to obtain feature data to be identified in different levels.

Optionally, in the application, more features are learned from the input sequence, and the accuracy and recall rate of named entity identification can be improved to a certain extent. The same input sequence can automatically learn features of different levels from different angles, and each module of multi-level feature extraction respectively combines an attention mechanism, a convolutional neural network, a long-term memory network and the like aiming at different extraction targets. The method is highlighted by using an attention mechanism as macroscopic weight adjustment and extracting specific features by other methods, or directly using the limited visual field of a convolutional neural network as the important attention of the current features and ignoring other information outside the visual field. The characteristics of different levels are fully excavated through the extensive attention mechanism, and the performance of a single characteristic extraction module can be improved to a great extent. In addition, the problem that the accuracy rate is reduced as the network deepens can be prevented by the residual error network.

Specifically, the method and the device can set the dimension of the output word vector and the minimum occurrence frequency of the words needing to be trained, store the generated word vector into a form that the text and the Chinese words form one-to-one correspondence, and search the word vector of the words in the sentence during the subsequent neural network training.

Further, the data is sorted into a word in each row, the word in each row corresponds to a word vector one by one, the entity category to which the word belongs is an output label, the label adopts one-hot coding, and the data is used as training data of a convolutional neural network, for example, the form is as follows: t { (x)₁,y₁),(x₂,y₂),...,(x_n,y_n) }; x represents a word sequence needing entity labeling, y represents an entity category label corresponding to the input word sequence, and the identified types comprise a person name, a place name and an organization name.

The convolution depth neural network is composed of an input layer, a hidden layer and an output layer, the weight of the neural network is initialized randomly, the output is calculated, and the optimal weight parameter, namely a convolution neural network model, is obtained through a back propagation algorithm and an Adam optimization iterator and through loop learning.

And S104, carrying out sequence marking on the characteristic data to be identified to obtain an entity identification result.

Further, named entity recognition can often be translated into a sequence tagging problem, where in the present application sequence tagging refers to a given input sequence: x ═ X₁,x₂,...,x_n]Each element is tagged with a certain tag in the tagged set: y ═ Y₁,y₂,...,y_n]. In this case, labeling is often performed by using a biees scheme, that is, a tag Set ═ B, I, E, O, S]B denotes start, I denotes middle, E denotes end, O denotes others, and S denotes a single entity. For example, New York represents a place entity, which may be labeled New/[ B-LOC ]]York/[E-LOC]. Thus, essentially every element in the input sequence is classified according to context, and the process of named entity identification is to identify the boundaries and classes of entities. The module extracts past and future information by utilizing a bidirectional long-term memory network, and solves the problem of long-term dependence. Meanwhile, in order to avoid mislabeling, a conditional random field is used to take adjacent labels as constraint conditions, for example, the label I-LOC is not likely to be followed by the label B-LOC (Softmax does not use the information), and finally, a high-quality sequence prediction result is obtained.

The application provides a named entity identification method fused with an breadth attention mechanism, which specifically comprises the following steps: based on the thought of extensive learning, more embedded expression modes can capture more various characteristics. Firstly, an input sequence is represented by embedding, and the embedding representation modes comprise embedding of pre-training words, embedding of random initialization characters and the like. On the basis, the multi-level feature extraction module is used for capturing features of different levels, including global character features (featureCG) and word features (featureWG), local character features (featureCL) and word features (featureWL). The feature extraction module respectively combines an attention mechanism, a convolutional neural network, a long-term memory network and the like aiming at different extraction targets. And then, carrying out feature fusion through weight self-adaptation, and obtaining a final prediction result through a sequence labeling module fusing a long-time memory network and a conditional random field.

In the application, after the data to be recognized is obtained, the data to be recognized can be converted into word embedding vectors based on a pre-trained word embedding vector matrix; performing multi-level feature extraction on the word embedding vector by using a convolutional neural network, a recursive neural network and an attention mechanism to obtain feature data to be identified in different levels; and carrying out sequence marking on the characteristic data to be identified to obtain an entity identification result. By applying the technical scheme of the application, compared with the prior art, breadth learning and attention mechanism are combined for named entity recognition. And then the effect of improving the identification accuracy can be achieved.

Optionally, the application may further perform label labeling on the feature data to be identified;

Further, the feature data to be identified at different levels in the present application at least includes any one of the following:

In addition, before the multi-level feature extraction is performed on the word embedding vector by using a convolutional neural network, a recurrent neural network and an attention mechanism to obtain feature data to be identified at different levels, the following steps can be further implemented:

Further, the present application may identify, through a neural network data classification model, a sample feature (for example, a name feature, a place name feature, a connecting word feature, and the like) of at least one object included in the sample data. Furthermore, the neural network data classification model may classify each sample feature in the sample data, and classify the sample features belonging to the same class into the same type, so that a plurality of sample features obtained after semantic segmentation of the sample data may be sample features composed of a plurality of different types.

It should be noted that, when the neural network data classification model performs semantic segmentation processing on sample data, the more accurate the classification of the pixel points in the sample data is, the higher the accuracy of identifying the mark object in the sample data is. It should be noted that the preset condition may be set by a user.

For example, the preset conditions may be set as: the classification accuracy rate of the feature classes reaches more than 70%, then, sample data repeatedly trains the neural network data classification model, and when the classification accuracy rate of the neural network data classification model on the feature classes reaches more than 70%, then the neural network data classification model can be applied to the embodiment of the application to perform semantic segmentation processing on the key feature data.

Optionally, for the used neural network data classification model, in an embodiment, the neural network data classification model may be trained through sample data. Specifically, sample data can be acquired, and the preset neural network data classification model is trained by using the sample data to obtain the neural network data classification model meeting the preset conditions.

In another embodiment of the present application, as shown in fig. 2, the present application further provides a named entity recognition apparatus, which includes an obtaining module 201, a converting module 202, an extracting module 203, and a generating module 204, where:

an obtaining module 201 configured to obtain data to be identified;

a conversion module 202 configured to convert the data to be recognized into word embedding vectors based on a pre-trained word embedding vector matrix;

the extraction module 203 is configured to perform multi-level feature extraction on the word embedding vector by using a convolutional neural network, a recurrent neural network and an attention mechanism to obtain feature data to be identified at different levels;

the generating module 204 is configured to perform sequence tagging on the feature data to be identified to obtain an entity identification result.

Optionally, in another embodiment of the present application, the method further includes an extracting module 203, where:

the extraction module 203 is configured to determine a weight parameter of each feature data to be identified based on the adjustment parameter;

and the extraction module 203 is configured to perform breadth feature fusion on each feature data to be identified based on the weight parameter of the feature data to be identified.

Optionally, in another embodiment of the present application, the generating module 204 further includes:

a generating module 204 configured to label the feature data to be identified;

a generating module 204 configured to classify each data in the feature data to be identified based on a biees labeling mode;

the generating module 204 is configured to perform sequence tagging on the feature data to be identified based on the tag tagging and the category in the feature data to be identified, so as to obtain the entity identification result.

a generating module 204 configured to obtain sample data, wherein the sample data includes at least one entity naming feature;

the generating module 204 is configured to train a preset convolutional neural network model by using the sample data to obtain the neural network meeting the preset condition and with the fusion breadth of attention.

Optionally, in another embodiment of the present application, the feature data to be identified at different levels at least includes any one of the following:

Fig. 3 is a block diagram of a logical structure of an electronic device. For example, the electronic device 300 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, an exercise device, a personal digital assistant, and the like.

Referring to fig. 3, electronic device 300 may include one or more of the following components: a processor 301 and a memory 302.

The processor 301 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and so on. The processor 301 may be implemented in at least one hardware form of a DSP (Digital Signal Processing), an FPGA (Field-Programmable Gate Array), and a PLA (Programmable Logic Array). The processor 301 may also include a main processor and a coprocessor, where the main processor is a processor for processing data in an awake state, and is also called a Central Processing Unit (CPU); a coprocessor is a low power processor for processing data in a standby state. In some embodiments, the processor 301 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content required to be displayed on the display screen. In some embodiments, the processor 301 may further include an AI (Artificial Intelligence) processor for processing computing operations related to machine learning.

Memory 302 may include one or more computer-readable storage media, which may be non-transitory. Memory 302 may also include high speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In some embodiments, a non-transitory computer readable storage medium in the memory 302 is configured to store at least one instruction for execution by the processor 301 to implement the interactive special effect calibration method provided by the method embodiments of the present application.

In some embodiments, the electronic device 300 may further include: a peripheral interface 303 and at least one peripheral. The processor 301, memory 302 and peripheral interface 303 may be connected by a bus or signal lines. Each peripheral may be connected to the peripheral interface 303 by a bus, signal line, or circuit board. Specifically, the peripheral device includes: at least one of radio frequency circuitry 304, touch display screen 305, camera 306, audio circuitry 307, positioning components 308, and power supply 309.

The peripheral interface 303 may be used to connect at least one peripheral related to I/O (Input/Output) to the processor 301 and the memory 302. In some embodiments, processor 301, memory 302, and peripheral interface 303 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 301, the memory 302 and the peripheral interface 303 may be implemented on a separate chip or circuit board, which is not limited by the embodiment.

The Radio Frequency circuit 304 is used for receiving and transmitting RF (Radio Frequency) signals, also called electromagnetic signals. The radio frequency circuitry 304 communicates with communication networks and other communication devices via electromagnetic signals. The rf circuit 304 converts an electrical signal into an electromagnetic signal to transmit, or converts a received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit 304 comprises: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a subscriber identity module card, and so forth. The radio frequency circuitry 304 may communicate with other terminals via at least one wireless communication protocol. The wireless communication protocols include, but are not limited to: metropolitan area networks, various generation mobile communication networks (2G, 3G, 4G, and 5G), Wireless local area networks, and/or WiFi (Wireless Fidelity) networks. In some embodiments, the rf circuit 304 may further include NFC (Near Field Communication) related circuits, which are not limited in this application.

The display screen 305 is used to display a UI (User Interface). The UI may include graphics, text, icons, video, and any combination thereof. When the display screen 305 is a touch display screen, the display screen 305 also has the ability to capture touch signals on or over the surface of the display screen 305. The touch signal may be input to the processor 301 as a control signal for processing. At this point, the display screen 305 may also be used to provide virtual buttons and/or a virtual keyboard, also referred to as soft buttons and/or a soft keyboard. In some embodiments, the display screen 305 may be one, providing the front panel of the electronic device 300; in other embodiments, the display screens 305 may be at least two, respectively disposed on different surfaces of the electronic device 300 or in a folded design; in some embodiments, the display screen 305 may be a flexible display screen disposed on a curved surface or a folded surface of the electronic device 300. Even further, the display screen 305 may be arranged in a non-rectangular irregular figure, i.e. a shaped screen. The Display screen 305 may be made of LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode), and the like.

The camera assembly 306 is used to capture images or video. Optionally, camera assembly 306 includes a front camera and a rear camera. Generally, a front camera is disposed at a front panel of the terminal, and a rear camera is disposed at a rear surface of the terminal. In some embodiments, the number of the rear cameras is at least two, and each rear camera is any one of a main camera, a depth-of-field camera, a wide-angle camera and a telephoto camera, so that the main camera and the depth-of-field camera are fused to realize a background blurring function, and the main camera and the wide-angle camera are fused to realize panoramic shooting and VR (Virtual Reality) shooting functions or other fusion shooting functions. In some embodiments, camera assembly 306 may also include a flash. The flash lamp can be a monochrome temperature flash lamp or a bicolor temperature flash lamp. The double-color-temperature flash lamp is a combination of a warm-light flash lamp and a cold-light flash lamp, and can be used for light compensation at different color temperatures.

Audio circuitry 307 may include a microphone and a speaker. The microphone is used for collecting sound waves of a user and the environment, converting the sound waves into electric signals, and inputting the electric signals to the processor 301 for processing or inputting the electric signals to the radio frequency circuit 304 to realize voice communication. For the purpose of stereo sound collection or noise reduction, a plurality of microphones may be provided at different portions of the electronic device 300. The microphone may also be an array microphone or an omni-directional pick-up microphone. The speaker is used to convert electrical signals from the processor 301 or the radio frequency circuitry 304 into sound waves. The loudspeaker can be a traditional film loudspeaker or a piezoelectric ceramic loudspeaker. When the speaker is a piezoelectric ceramic speaker, the speaker can be used for purposes such as converting an electric signal into a sound wave audible to a human being, or converting an electric signal into a sound wave inaudible to a human being to measure a distance. In some embodiments, audio circuitry 307 may also include a headphone jack.

The positioning component 308 is used to locate the current geographic location of the electronic device 300 to implement navigation or LBS (location based Service). The positioning component 308 may be a positioning component based on the GPS (global positioning System) in the united states, the beidou System in china, the graves System in russia, or the galileo System in the european union.

The power supply 309 is used to supply power to various components in the electronic device 300. The power source 309 may be alternating current, direct current, disposable batteries, or rechargeable batteries. When the power source 309 includes a rechargeable battery, the rechargeable battery may support wired or wireless charging. The rechargeable battery may also be used to support fast charge technology.

In some embodiments, the electronic device 300 also includes one or more sensors 310. The one or more sensors 310 include, but are not limited to: acceleration sensor 311, gyro sensor 312, pressure sensor 313, fingerprint sensor 314, optical sensor 315, and proximity sensor 316.

The acceleration sensor 311 may detect the magnitude of acceleration in three coordinate axes of a coordinate system established with the electronic device 300. For example, the acceleration sensor 311 may be used to detect components of the gravitational acceleration in three coordinate axes. The processor 301 may control the touch display screen 305 to display the user interface in a landscape view or a portrait view according to the gravitational acceleration signal collected by the acceleration sensor 311. The acceleration sensor 311 may also be used for acquisition of motion data of a game or a user.

The gyro sensor 312 may detect a body direction and a rotation angle of the electronic device 300, and the gyro sensor 312 and the acceleration sensor 311 may cooperate to acquire a 3D motion of the user on the electronic device 300. The processor 301 may implement the following functions according to the data collected by the gyro sensor 312: motion sensing (such as changing the UI according to a user's tilting operation), image stabilization at the time of photographing, game control, and inertial navigation.

The pressure sensors 313 may be disposed on a side bezel of the electronic device 300 and/or an underlying layer of the touch display screen 305. When the pressure sensor 313 is arranged on the side frame of the electronic device 300, the holding signal of the user to the electronic device 300 can be detected, and the processor 301 performs left-right hand recognition or shortcut operation according to the holding signal collected by the pressure sensor 313. When the pressure sensor 313 is disposed at the lower layer of the touch display screen 305, the processor 301 controls the operability control on the UI interface according to the pressure operation of the user on the touch display screen 305. The operability control comprises at least one of a button control, a scroll bar control, an icon control and a menu control.

The fingerprint sensor 314 is used for collecting a fingerprint of the user, and the processor 301 identifies the identity of the user according to the fingerprint collected by the fingerprint sensor 314, or the fingerprint sensor 314 identifies the identity of the user according to the collected fingerprint. Upon identifying that the user's identity is a trusted identity, processor 301 authorizes the user to perform relevant sensitive operations including unlocking the screen, viewing encrypted information, downloading software, paying, and changing settings, etc. The fingerprint sensor 314 may be disposed on the front, back, or side of the electronic device 300. When a physical button or vendor Logo is provided on the electronic device 300, the fingerprint sensor 314 may be integrated with the physical button or vendor Logo.

The optical sensor 315 is used to collect the ambient light intensity. In one embodiment, the processor 301 may control the display brightness of the touch screen display 305 based on the ambient light intensity collected by the optical sensor 315. Specifically, when the ambient light intensity is high, the display brightness of the touch display screen 305 is increased; when the ambient light intensity is low, the display brightness of the touch display screen 305 is turned down. In another embodiment, the processor 301 may also dynamically adjust the shooting parameters of the camera head assembly 306 according to the ambient light intensity collected by the optical sensor 315.

The proximity sensor 316, also referred to as a distance sensor, is typically disposed on the front panel of the electronic device 300. The proximity sensor 316 is used to capture the distance between the user and the front of the electronic device 300. In one embodiment, the processor 301 controls the touch display screen 305 to switch from the bright screen state to the dark screen state when the proximity sensor 316 detects that the distance between the user and the front surface of the electronic device 300 gradually decreases; when the proximity sensor 316 detects that the distance between the user and the front surface of the electronic device 300 is gradually increased, the processor 301 controls the touch display screen 305 to switch from the breath screen state to the bright screen state.

Those skilled in the art will appreciate that the configuration shown in fig. 3 is not intended to be limiting of electronic device 300, and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components may be used.

In an exemplary embodiment, there is also provided a non-transitory computer-readable storage medium, such as the memory 304, comprising instructions executable by the processor 320 of the electronic device 300 to perform the named entity identification method described above, the method comprising: acquiring data to be identified; converting the data to be recognized into word embedding vectors based on a pre-trained word embedding vector matrix; performing multi-level feature extraction on the word embedding vector by using a convolutional neural network, a recursive neural network and an attention mechanism to obtain feature data to be identified in different levels; and carrying out sequence marking on the characteristic data to be identified to obtain an entity identification result. Optionally, the instructions may also be executable by the processor 320 of the electronic device 300 to perform other steps involved in the exemplary embodiments described above. For example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

In an exemplary embodiment, there is also provided an application/computer program product comprising one or more instructions executable by the processor 320 of the electronic device 300 to perform the named entity recognition method described above, the method comprising: acquiring data to be identified; converting the data to be recognized into word embedding vectors based on a pre-trained word embedding vector matrix; performing multi-level feature extraction on the word embedding vector by using a convolutional neural network, a recursive neural network and an attention mechanism to obtain feature data to be identified in different levels; and carrying out sequence marking on the characteristic data to be identified to obtain an entity identification result. Optionally, the instructions may also be executable by the processor 320 of the electronic device 300 to perform other steps involved in the exemplary embodiments described above.

Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.

It will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims

1. A named entity recognition method, comprising:

acquiring data to be identified;

2. The method of claim 1, wherein after the performing multi-level feature extraction on the word embedding vector by using a convolutional neural network, a recurrent neural network and an attention mechanism to obtain different levels of feature data to be recognized, the method further comprises:

3. The method of claim 1, wherein the performing sequence labeling on the feature data to be recognized to obtain an entity recognition result comprises:

labeling the characteristic data to be identified with a label;

4. The method of claim 2, wherein the different levels of feature data to be identified comprise at least any one of:

5. The method of claim 1, wherein before the performing multi-level feature extraction on the word embedding vector by using a convolutional neural network, a recurrent neural network and an attention mechanism to obtain different levels of feature data to be recognized, the method further comprises:

and training a preset neural network model by using the sample data to obtain the neural network meeting the fusion breadth attention of the preset condition.

6. A named entity recognition apparatus, comprising:

the acquisition module is configured to acquire data to be identified;

7. An electronic device, comprising:

a display for display with the memory to execute the executable instructions to perform the operations of the named entity recognition method of any of claims 1-5.

8. A computer-readable storage medium storing computer-readable instructions that, when executed, perform the operations of the named entity recognition method of any of claims 1-5.