Movatterモバイル変換


[0]ホーム

URL:


CN109165384A - A kind of name entity recognition method and device - Google Patents

A kind of name entity recognition method and device
Download PDF

Info

Publication number
CN109165384A
CN109165384ACN201810971954.5ACN201810971954ACN109165384ACN 109165384 ACN109165384 ACN 109165384ACN 201810971954 ACN201810971954 ACN 201810971954ACN 109165384 ACN109165384 ACN 109165384A
Authority
CN
China
Prior art keywords
text
word
recognized
vector
semantic information
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201810971954.5A
Other languages
Chinese (zh)
Inventor
王伟才
王纯斌
赵神州
覃进学
王升升
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chengdu Sefon Software Co Ltd
Original Assignee
Chengdu Sefon Software Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chengdu Sefon Software Co LtdfiledCriticalChengdu Sefon Software Co Ltd
Priority to CN201810971954.5ApriorityCriticalpatent/CN109165384A/en
Publication of CN109165384ApublicationCriticalpatent/CN109165384A/en
Pendinglegal-statusCriticalCurrent

Links

Classifications

Landscapes

Abstract

The embodiment of the present invention provides a kind of name entity recognition method and device.Method includes obtaining text to be identified;The corresponding phonetic vector of each word is obtained by the corresponding word vector of word each in text input word incorporation model to be identified acquisition text to be identified, and by text input phonetic incorporation model to be identified;The corresponding phonetic vector of each word vector sum is combined acquisition mix vector, semantic coding will be carried out in all combinatorics on words vector input BiLSTM, obtain the corresponding semantic information feature of text to be identified;Corresponding entity tag sequence in text to be identified is obtained according to semantic information feature.Device is for executing the above method, the embodiment of the present invention is by respectively obtaining the word vector sum phonetic vector of text to be identified according to word incorporation model and phonetic vector model, and word vector sum phonetic vector is combined to be input in BiLSTM and is identified, so as to make up the deficiency of word vector expression well, so that the accuracy rate of identification greatly improves.

Description

Named entity identification method and device
Technical Field
The invention relates to the technical field of language processing, in particular to a named entity identification method and device.
Background
Named Entity Recognition (NER) is a task based on natural language processing, and is used for recognizing a Named term from a text and laying down tasks such as relation extraction. In the narrow sense, proper nouns such as names of people, places, organizations and the like are recognized. In a particular domain, various entity types within the domain may be defined accordingly. Named entity identification is an important basic tool in application fields such as information extraction, question and answer systems, syntactic analysis, machine translation, Semantic Web-oriented metadata labeling and the like, and plays an important role in the process of bringing natural language processing technology into practical use.
The existing named entity recognition method is mainly divided into three categories: the first type is a rule-based approach that identifies named entities by matching text to rules using manually written rules. The second category is a feature template-based method, in which a statistical machine learning method considers named entity recognition as a sequence labeling task, and a large-scale corpus is used to learn a labeling model, so as to label each position of a sentence. The third category is neural network-based methods, which, with the advent of distributed representation of words, become models that can efficiently handle many NLP sequence tagging tasks. The method makes the training of the model an end-to-end integral process, is independent of characteristic engineering, is a data-driven method, and has not high accuracy when identifying the Chinese.
Disclosure of Invention
In view of the above, an object of the embodiments of the present invention is to provide a method and an apparatus for identifying a named entity, so as to solve the above technical problems.
In a first aspect, an embodiment of the present invention provides a method for identifying a named entity, including:
acquiring a text to be recognized, wherein the text to be recognized comprises a plurality of words;
embedding the text input words to be recognized into a model to obtain a word vector corresponding to each word in the text to be recognized, and inputting the text to be recognized into a pinyin embedded model to obtain a pinyin vector corresponding to each word;
combining each word vector and the corresponding pinyin vector to obtain a combined vector, inputting the combined vectors respectively corresponding to all the words into a bidirectional long-short time memory network (BilTM) for semantic coding, and obtaining semantic information characteristics corresponding to the text to be recognized;
and obtaining a corresponding entity tag sequence in the text to be recognized according to the semantic information characteristics.
Further, the obtaining a corresponding entity tag sequence in the text to be recognized according to the semantic information features includes:
normalizing the semantic information features by utilizing a softmax function to obtain the emission probability value of each word in the text to be recognized as each label;
and obtaining the label corresponding to the maximum emission probability value corresponding to each word, and forming the entity label sequence according to the labels corresponding to all the words.
Further, the embedding the text input word to be recognized into a model to obtain a word vector corresponding to each word in the text to be recognized includes:
and inputting the text to be recognized into the word embedding model, and performing OneHot coding mapping on each word by the word embedding model to obtain a word vector corresponding to each word, wherein the length of the word vector is fixed.
Further, the BilSTM comprises a forward long-short time memory network LSTM and a reverse LSTM; correspondingly, the step of inputting the combined vector into a bidirectional long-and-short time memory network BilSTM for semantic coding to obtain semantic information characteristics corresponding to the text to be recognized includes:
the forward SLTM extracts features of the text to be recognized to obtain a first hidden state sequence;
the reverse SLTM performs feature extraction on the text to be recognized to obtain a second hidden state sequence;
and splicing the first hidden state sequence and the second hidden state sequence according to the sequence of each word in the text to be recognized to obtain the semantic information characteristics.
Further, the obtaining a corresponding entity tag sequence in the text to be recognized according to the semantic information features includes:
normalizing the semantic information by utilizing a softmax function to obtain the emission probability value of each word in the text to be recognized as each label;
and inputting the emission probability value into a conditional random field CRF, and acquiring the entity label sequence according to the emission probability value of each label and a transition probability matrix in the CRF.
Further, the inputting the emission probability value into a conditional random field CRF, and obtaining the entity tag sequence according to the emission probability value of each tag and a transition probability matrix in the CRF includes:
according toCalculating and obtaining the total probability value of the text to be recognized;
according toNormalizing the total probability value to obtain the entity label sequence; wherein,the transmission probability value corresponding to the ith character in the text to be recognized, n is the number of the characters of the text to be recognized,the transfer probability value corresponding to the transmission probability value of the ith word in the text to be recognized is obtained; y isxN and i are positive integers, and i is less than or equal to n.
Further, the method further comprises:
and constructing the pinyin embedding model through a convolutional neural network and a maximum pooling method.
In a second aspect, an embodiment of the present invention provides a named entity identifying device, including:
the device comprises an acquisition module, a recognition module and a recognition module, wherein the acquisition module is used for acquiring a text to be recognized, and the text to be recognized comprises a plurality of characters;
the vector identification module is used for embedding the text input words to be identified into a model to obtain a word vector corresponding to each word in the text to be identified, and inputting the text to be identified into a pinyin embedding model to obtain a pinyin vector corresponding to each word;
the semantic coding module is used for combining each word vector and the corresponding pinyin vector to obtain a combined vector, inputting the combined vectors corresponding to all the words into a bidirectional long-short time memory network (BilTM) for semantic coding, and obtaining semantic information characteristics corresponding to the text to be recognized;
and the marking module is used for obtaining the corresponding entity label sequence in the text to be recognized according to the semantic information characteristics.
In a third aspect, an embodiment of the present invention provides an electronic device, including: a processor, a memory, and a bus, wherein,
the processor and the memory are communicated with each other through the bus;
the memory stores program instructions executable by the processor, the processor being capable of performing the method steps of the first aspect when invoked by the program instructions.
In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, including:
the non-transitory computer readable storage medium stores computer instructions that cause the computer to perform the method steps of the first aspect.
According to the embodiment of the invention, the word vector and the pinyin vector corresponding to the text to be recognized are respectively obtained according to the word embedding model and the pinyin vector model, and the word vector and the pinyin vector are combined and input into the BilSTM for recognition, so that the defect of word vector representation can be well compensated, and the recognition accuracy is greatly improved.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a schematic flow chart of a named entity identification method according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating a named entity recognition model according to an embodiment of the present invention;
FIG. 3 is a diagram illustrating a Pinyin embedding model according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a named entity recognition apparatus according to an embodiment of the present invention;
fig. 5 is a block diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
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, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Fig. 1 is a schematic flow chart of a named entity identification method according to an embodiment of the present invention, as shown in fig. 1, the method includes:
step 101: acquiring a text to be recognized, wherein the text to be recognized comprises a plurality of words;
in a specific implementation process, the recognition device first obtains a text to be recognized, which should be noted that the text to be recognized may be a section of text to be recognized, and the text includes a plurality of chinese characters. The named entity recognition is to recognize words belonging to a preset tag in the text, where the preset tag may be: the name of a person, a place, a government, or a weapon may be defined in advance according to actual circumstances. The embodiment of the invention adopts a label set of BIEO, B represents the beginning (Begin) of an entity, I represents the middle (Intermediate) of the entity, E represents the End (End) of the entity, and O represents a non-entity (Other). For example, if the entity identification of the name is to be performed, taking "learning the mine peak good list" as an example, the tagging sequence is "oobooo", and thus the name "mine peak" is extracted.
Step 102: embedding the text input words to be recognized into a model to obtain a word vector corresponding to each word in the text to be recognized, and inputting the text to be recognized into a pinyin embedded model to obtain a pinyin vector corresponding to each word;
in the specific implementation process, the device comprises an identification model, wherein the identification model comprises a word embedding model, a pinyin embedding model and a BilSTM network, and also comprises a CRF layer, wherein the word embedding model and the pinyin embedding model form an embedding layer, and the BilSTM network forms a BilSTM layer. The device firstly inputs a text to be recognized into a word embedding model and a pinyin embedding model, the word embedding model generates a word vector corresponding to each word in the text to be recognized, and the pinyin embedding model generates a pinyin vector corresponding to each word in the text to be recognized. It should be noted that the working principle of the word embedding model is to perform One Hot encoding mapping on each word in the received text to be recognized, so as to obtain a low-dimensional dense word vector, and the length of the word vector is fixed. Moreover, the length of the pinyin vector is also fixed, and if the length of the pinyin vector of each word obtained by the pinyin embedding model is smaller than the set length, the pinyin vectors can be complemented by PADDING, such as: < PADDING, W, a, N, G, PADDING >, etc.
Step 103: combining each word vector and the corresponding pinyin vector to obtain a combined vector, inputting the combined vectors respectively corresponding to all the words into a bidirectional long-short time memory network (BilTM) for semantic coding, and obtaining semantic information characteristics corresponding to the text to be recognized;
in a specific implementation process, after the word vector and the pinyin vector corresponding to each word are obtained, the word vector and the pinyin vector are connected to obtain a combined vector, specifically, the pinyin vector is connected to the back of the word vector. Assuming that the length of the word vector is p and the length of the pinyin vector is q, the length of the embedded model layer is w ═ p | | | q. Where | represents vector concatenation. And taking the combined vector as the input of the BilSTM, and performing semantic coding in the BilSTM so as to obtain the semantic information characteristics corresponding to the text to be recognized. It should be noted that the semantic information feature refers to a score of each word in the text to be recognized as each label, which is not normalized.
Step 104: and obtaining a corresponding entity tag sequence in the text to be recognized according to the semantic information characteristics.
In a specific implementation process, after the device acquires the semantic information features, for each word, the device acquires the label with the highest score as the label of the word, so that the corresponding entity label sequence in the text to be recognized can be acquired.
According to the embodiment of the invention, the word vector and the pinyin vector corresponding to the text to be recognized are respectively obtained according to the word embedding model and the pinyin vector model, and the word vector and the pinyin vector are combined and input into the BilSTM for recognition, so that the defect of word vector representation can be well compensated, and the recognition accuracy is greatly improved.
On the basis of the above embodiment, the obtaining a corresponding entity tag sequence in the text to be recognized according to the semantic information features includes:
normalizing the semantic information features by utilizing a softmax function to obtain the emission probability value of each word in the text to be recognized as each label;
and obtaining the label corresponding to the maximum emission probability value corresponding to each word, and forming the entity label sequence according to the labels corresponding to all the words.
In a specific implementation process, normalization is carried out by utilizing a softmax function according to the obtained semantic feature information, the emission probability value that each word in the text to be recognized is each label is obtained, and the score corresponding to each word in the semantic feature information is enabled to be between [0,1 ]. For each word, the label with the highest emission probability value is used as the label of the word, and the labels corresponding to all the words in the text to be recognized form an entity label sequence. It should be noted that the label with the highest score can also be directly taken as the label of the word from the semantic information features, and then the labels of all words form the entity label sequence.
On the basis of the above embodiment, the BiLSTM includes a forward long-short time memory network LSTM and a reverse LSTM; correspondingly, the step of inputting the combined vector into a bidirectional long-and-short time memory network BilSTM for semantic coding to obtain semantic information characteristics corresponding to the text to be recognized includes:
the forward SLTM extracts features of the text to be recognized to obtain a first hidden state sequence; the reverse SLTM performs feature extraction on the text to be recognized to obtain a second hidden state sequence; and splicing the first hidden state sequence and the second hidden state sequence according to the sequence of each word in the text to be recognized to obtain the semantic information characteristics.
In a specific implementation, the BilSTM comprises a forward long-short-term memory network LSTM and a backward LSTM, and a vector (x) is combined1,x2,…,xn) After the BiLSTM is input, the forward LSTM operates the combined vector to obtain a first hidden state sequenceThe reverse LSTM operates the combination vector to obtain a second hidden state sequenceWill be provided withAndthe hidden states output at all positions are spliced according to the positions to obtain the semantic information characteristicsIt should be noted that the position refers to the position of each word in the text to be recognized.
On the basis of the above embodiment, the obtaining a corresponding entity tag sequence in the text to be recognized according to the semantic information features includes:
normalizing the semantic information by utilizing a softmax function to obtain the emission probability value of each word in the text to be recognized as each label; and inputting the emission probability value into a conditional random field CRF, and acquiring the entity label sequence according to the emission probability value of each label and a transition probability matrix in the CRF.
In a specific implementation process, theThe recognition result of the text to be recognized obtained by BilSTM may be invalid, for example: when the label of the two continuous words is B, the identification is invalid. At this time, normalization can be performed by utilizing a softmax function according to the obtained semantic feature information, the emission probability value that each word in the text to be recognized is each label is obtained, and the score corresponding to each word in the semantic feature information is all [0,1]]In the meantime. Inputting the emission probability value into a CRF, wherein the parameter included in the CRF is a matrix A of (k +2) × (k +2), k is the number of preset labels, AijThe transition score from the ith tag to the jth tag is shown, plus 2 because a start state is added for the beginning of the sentence and an end state is added for the end of the sentence. The label sequence corresponding to the length of the text to be recognized is marked as y ═ y (y)1,y2,...,yn) Then the CRF model scores y for a label of the text x to be recognized asThe total probability value of the whole text to be recognized is equal to the sum of scores of all positions, and the score of each position consists of two parts, wherein one part is p output by BilSTMiDetermining where p isiIs based on semantic information characteristics obtained by BilSTMThe other part is determined by the state transition matrix A of the CRF through Softmax calculation. Normalized by Softmax, has a probability ofWherein Y isxAll tag sequences are meant to include both possible and impossible tag sequences. Wherein,is the transmission probability value corresponding to the ith character in the text to be recognized, n is the number of the characters of the text to be recognized,a transfer probability value corresponding to the transmission probability value of the ith word in the text to be recognized; y isxN and i are positive integers, and i is less than or equal to n.
When the CRF is trained in advance, the CRF model is trained by maximizing the log-likelihood function, and the model uses a dynamically-programmed Viterbi algorithm to solve the optimal path in prediction.
The embodiment of the invention inputs the emission probability value into the CRF for labeling, thereby solving the problem of invalid output of the BilSTM and further improving the identification accuracy.
FIG. 2 is a schematic diagram of a named entity recognition model provided by an embodiment of the present invention, as shown in FIG. 2, the model includes a word embedding model, a pinyin embedding model, a BilSTM model and a CRF layer, wherein the BilSTM model includes a forward LSTM and a backward LSTM, taking the text to be recognized as "I love Chinese", firstly embedding the input word of "I love Chinese" into the model to obtain a word vector of each word, then inputting "I love Chinese" into the pinyin embedding model to obtain a pinyin vector of each word, combining the word vector and the pinyin vector of each word to obtain a combined vector, then inputting the combined vector into the forward LSTM and the backward LSTM to obtain a first hidden state sequence and a second hidden state sequence respectively, splicing the first hidden state sequence and the second hidden state sequence to obtain semantic information characteristics, then inputting the semantic information characteristics into the CRF to realize labeling of the recognized text, obtaining the entity tag sequence.
On the basis of the above embodiment, the method further includes:
and constructing the pinyin embedding model through a convolutional neural network and a maximum pooling method.
In the specific implementation process, before the identification, a pinyin embedding model needs to be constructed in advance,
fig. 3 is a schematic diagram of a pinyin embedding model provided in an embodiment of the present invention, and as shown in fig. 3, taking a "network" as an example, where the pinyin of the "network" is "wanglo", and since the length of "wanglo" is smaller than a preset length, Padding is used for both front and back sides, then "wanglo" is input into a word embedding model, convolution calculation is performed on "wanglo" through a convolution layer, in order to avoid overfitting of the convolution neural network, a pooling layer is added behind a volume base layer, and a pinyin vector is obtained through maximum pooling.
The embodiment of the invention constructs the pinyin embedded model through the convolutional neural network and the maximum pooling method, prevents the convolutional neural network from being over-fitted, and thus obtains a proper pinyin vector.
Fig. 4 is a schematic structural diagram of a named entity recognition apparatus according to an embodiment of the present invention, as shown in fig. 4, the apparatus includes: an acquisition module 401, a vector recognition module 402, a semantic coding module 403, and a labeling module 404, wherein,
the obtaining module 401 is configured to obtain a text to be recognized, where the text to be recognized includes multiple words; the vector identification module 402 is configured to embed the text input word to be identified into a model to obtain a word vector corresponding to each word in the text to be identified, and to input the text to be identified into a pinyin embedding model to obtain a pinyin vector corresponding to each word; the semantic coding module 403 is configured to combine each word vector and the corresponding pinyin vector to obtain a combined vector, input the combined vectors corresponding to all the words into a bidirectional long-short time memory network BiLSTM for semantic coding, and obtain semantic information features corresponding to the text to be recognized; the labeling module 404 is configured to obtain a corresponding entity tag sequence in the text to be recognized according to the semantic information features.
On the basis of the foregoing embodiment, the labeling module 404 is specifically configured to:
normalizing the semantic information features by utilizing a softmax function to obtain the emission probability value of each word in the text to be recognized as each label;
and obtaining the label corresponding to the maximum emission probability value corresponding to each word, and forming the entity label sequence according to the labels corresponding to all the words.
On the basis of the foregoing embodiment, the vector identification module 402 is specifically configured to:
and inputting the text to be recognized into the word embedding model, and performing OneHot coding mapping on each word by the word embedding model to obtain a word vector corresponding to each word, wherein the length of the word vector is fixed.
On the basis of the above embodiment, the BiLSTM includes a forward long-short time memory network LSTM and a reverse LSTM; correspondingly, the semantic coding module 403 is specifically configured to:
the forward SLTM extracts features of the text to be recognized to obtain a first hidden state sequence;
the reverse SLTM performs feature extraction on the text to be recognized to obtain a second hidden state sequence;
and splicing the first hidden state sequence and the second hidden state sequence according to the sequence of each word in the text to be recognized to obtain the semantic information characteristics.
On the basis of the foregoing embodiment, the labeling module 404 is specifically configured to:
normalizing the semantic information by utilizing a softmax function to obtain the emission probability value of each word in the text to be recognized as each label;
and inputting the emission probability value into a conditional random field CRF, and acquiring the entity label sequence according to the emission probability value of each label and a transition probability matrix in the CRF.
On the basis of the foregoing embodiment, the labeling module 404 is specifically configured to:
according toComputingObtaining a total probability value of the text to be recognized;
according toNormalizing the total probability value to obtain the entity label sequence; wherein,the transmission probability value corresponding to the ith character in the text to be recognized, n is the number of the characters of the text to be recognized,the transfer probability value corresponding to the transmission probability value of the ith word in the text to be recognized is obtained; y isxN and i are positive integers, and i is less than or equal to n.
On the basis of the above embodiment, the apparatus further includes:
and the pinyin model establishing module is used for establishing the pinyin embedded model through a convolutional neural network and a maximum pooling method.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working process of the apparatus described above may refer to the corresponding process in the foregoing method, and will not be described in too much detail herein.
In summary, the embodiment of the invention obtains the word vector and the pinyin vector corresponding to the text to be recognized respectively according to the word embedding model and the pinyin vector model, and combines and inputs the word vector and the pinyin vector into the BilSTM for recognition, thereby well compensating the deficiency of word vector representation and greatly improving the recognition accuracy.
The present embodiment discloses a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method provided by the above-mentioned method embodiments, for example, comprising: acquiring a text to be recognized, wherein the text to be recognized comprises a plurality of words; embedding the text input words to be recognized into a model to obtain a word vector corresponding to each word in the text to be recognized, and inputting the text to be recognized into a pinyin embedded model to obtain a pinyin vector corresponding to each word; combining each word vector and the corresponding pinyin vector to obtain a combined vector, inputting the combined vectors respectively corresponding to all the words into a bidirectional long-short time memory network (BilTM) for semantic coding, and obtaining semantic information characteristics corresponding to the text to be recognized; and obtaining a corresponding entity tag sequence in the text to be recognized according to the semantic information characteristics.
The present embodiments provide a non-transitory computer-readable storage medium storing computer instructions that cause the computer to perform the methods provided by the above method embodiments, for example, including: acquiring a text to be recognized, wherein the text to be recognized comprises a plurality of words; embedding the text input words to be recognized into a model to obtain a word vector corresponding to each word in the text to be recognized, and inputting the text to be recognized into a pinyin embedded model to obtain a pinyin vector corresponding to each word; combining each word vector and the corresponding pinyin vector to obtain a combined vector, inputting the combined vectors respectively corresponding to all the words into a bidirectional long-short time memory network (BilTM) for semantic coding, and obtaining semantic information characteristics corresponding to the text to be recognized; and obtaining a corresponding entity tag sequence in the text to be recognized according to the semantic information characteristics.
Referring to fig. 5, fig. 5 is a block diagram of an electronic device according to an embodiment of the present invention. The electronic device may include a recognition model 501, a memory 502, a storage controller 503, a processor 504, a peripheral interface 505, an input output unit 506, an audio unit 507, a display unit 508.
The memory 502, the memory controller 503, the processor 504, the peripheral interface 505, the input/output unit 506, the audio unit 507, and the display unit 508 are electrically connected to each other directly or indirectly, so as to realize data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The recognition model 501 includes at least one software function module which may be stored in the memory 502 in the form of software or firmware (firmware) or solidified in an Operating System (OS) of the recognition model 501. The processor 504 is adapted to execute executable modules stored in the memory 502, such as software functional modules or computer programs comprised by the recognition model 501.
The Memory 502 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. The memory 502 is used for storing a program, and the processor 504 executes the program after receiving an execution instruction, and the method executed by the server defined by the flow process disclosed in any of the foregoing embodiments of the present invention may be applied to the processor 504, or implemented by the processor 504.
The processor 504 may be an integrated circuit chip having signal processing capabilities. The Processor 504 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components. The various methods, steps and logic blocks disclosed in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor 504 may be any conventional processor or the like.
The peripheral interface 505 couples various input/output devices to the processor 504 and to the memory 502. In some embodiments, the peripheral interface 505, the processor 504, and the memory controller 503 may be implemented in a single chip. In other examples, they may be implemented separately from the individual chips.
The input and output unit 506 is used for providing input data for a user to realize the interaction of the user with the server (or the local terminal). The input/output unit 506 may be, but is not limited to, a mouse, a keyboard, and the like.
Audio unit 507 provides an audio interface to a user, which may include one or more microphones, one or more speakers, and audio circuitry.
The display unit 508 provides an interactive interface (e.g., a user interface) between the electronic device and a user or for displaying image data to a user reference. In this embodiment, the display unit 508 may be a liquid crystal display or a touch display. In the case of a touch display, the display can be a capacitive touch screen or a resistive touch screen, which supports single-point and multi-point touch operations. Supporting single-point and multi-point touch operations means that the touch display can sense touch operations from one or more locations on the touch display at the same time, and the sensed touch operations are sent to the processor 504 for calculation and processing.
The peripheral interface 505 couples various input/output devices to the processor 504 and to the memory 502. In some embodiments, the peripheral interface 505, the processor 504, and the memory controller 503 may be implemented in a single chip. In other examples, they may be implemented separately from the individual chips.
The input/output unit 506 is used for providing input data for a user to realize the interaction of the user and the processing terminal. The input/output unit 506 may be, but is not limited to, a mouse, a keyboard, and the like.
It will be appreciated that the configuration shown in fig. 5 is merely illustrative and that the electronic device may include more or fewer components than shown in fig. 5 or may have a different configuration than shown in fig. 5. The components shown in fig. 5 may be implemented in hardware, software, or a combination thereof.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, the functional modules in the embodiments of the present invention may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention. 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, it need not be further defined and explained in subsequent figures.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

Claims (10)

CN201810971954.5A2018-08-232018-08-23A kind of name entity recognition method and devicePendingCN109165384A (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN201810971954.5ACN109165384A (en)2018-08-232018-08-23A kind of name entity recognition method and device

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN201810971954.5ACN109165384A (en)2018-08-232018-08-23A kind of name entity recognition method and device

Publications (1)

Publication NumberPublication Date
CN109165384Atrue CN109165384A (en)2019-01-08

Family

ID=64896659

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN201810971954.5APendingCN109165384A (en)2018-08-232018-08-23A kind of name entity recognition method and device

Country Status (1)

CountryLink
CN (1)CN109165384A (en)

Cited By (50)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN109858039A (en)*2019-03-012019-06-07北京奇艺世纪科技有限公司A kind of text information identification method and identification device
CN109871535A (en)*2019-01-162019-06-11四川大学 A French Named Entity Recognition Method Based on Deep Neural Network
CN109918681A (en)*2019-03-292019-06-21哈尔滨理工大学 A Semantic Matching Method for Fusion Problems Based on Chinese Characters and Pinyin
CN109960728A (en)*2019-03-112019-07-02北京市科学技术情报研究所(北京市科学技术信息中心)A kind of open field conferencing information name entity recognition method and system
CN109993227A (en)*2019-03-292019-07-09京东方科技集团股份有限公司 Method, system, device and medium for automatically adding International Classification of Diseases codes
CN110069781A (en)*2019-04-242019-07-30北京奇艺世纪科技有限公司A kind of recognition methods of entity tag and relevant device
CN110210037A (en)*2019-06-122019-09-06四川大学Category detection method towards evidence-based medicine EBM field
CN110298019A (en)*2019-05-202019-10-01平安科技(深圳)有限公司Name entity recognition method, device, equipment and computer readable storage medium
CN110348021A (en)*2019-07-172019-10-18湖北亿咖通科技有限公司Character string identification method, electronic equipment, storage medium based on name physical model
CN110413743A (en)*2019-08-092019-11-05安徽科大讯飞医疗信息技术有限公司A kind of key message abstracting method, device, equipment and storage medium
CN110619877A (en)*2019-09-272019-12-27安徽咪鼠科技有限公司Voice recognition man-machine interaction method, device and system applied to laser pen and storage medium
CN110782002A (en)*2019-09-122020-02-11成都四方伟业软件股份有限公司LSTM neural network training method and device
CN110827831A (en)*2019-11-152020-02-21广州洪荒智能科技有限公司Voice information processing method, device, equipment and medium based on man-machine interaction
CN110929749A (en)*2019-10-152020-03-27平安科技(深圳)有限公司Text recognition method, text recognition device, text recognition medium and electronic equipment
CN111104800A (en)*2019-12-242020-05-05东软集团股份有限公司Entity identification method, device, equipment, storage medium and program product
CN111125386A (en)*2019-12-022020-05-08腾讯科技(北京)有限公司Media resource processing method and device, storage medium and electronic device
CN111191459A (en)*2019-12-252020-05-22医渡云(北京)技术有限公司Text processing method and device, readable medium and electronic equipment
CN111222335A (en)*2019-11-272020-06-02上海眼控科技股份有限公司Corpus correction method and device, computer equipment and computer-readable storage medium
CN111339764A (en)*2019-09-182020-06-26华为技术有限公司 A Chinese named entity recognition method and device
CN111435411A (en)*2019-01-152020-07-21菜鸟智能物流控股有限公司Named body type identification method and device and electronic equipment
CN111444719A (en)*2020-03-172020-07-24车智互联(北京)科技有限公司Entity identification method and device and computing equipment
CN111507109A (en)*2019-01-142020-08-07阿里巴巴集团控股有限公司Named entity identification method and device of electronic medical record
CN111680513A (en)*2020-05-292020-09-18平安银行股份有限公司Feature information identification method and device and computer readable storage medium
CN111695385A (en)*2019-03-152020-09-22杭州海康威视数字技术股份有限公司Text recognition method, device and equipment
WO2020192237A1 (en)*2019-03-222020-10-01北京京东尚科信息技术有限公司Semantic recognition method, device and system based on artificial intelligence, and storage medium
CN111738808A (en)*2020-07-242020-10-02浙江口碑网络技术有限公司Data processing method, device and equipment
CN111831818A (en)*2019-04-152020-10-27北京京东尚科信息技术有限公司 Text information identification method, device and storage medium
CN111859963A (en)*2019-04-082020-10-30中移(苏州)软件技术有限公司 Named entity identification method, apparatus, apparatus and computer-readable storage medium
CN111859994A (en)*2020-06-082020-10-30北京百度网讯科技有限公司 Machine translation model acquisition and text translation method, device and storage medium
CN111967264A (en)*2020-08-262020-11-20湖北亿咖通科技有限公司Named entity identification method
CN111985235A (en)*2019-05-232020-11-24北京地平线机器人技术研发有限公司Text processing method and device, computer readable storage medium and electronic equipment
CN112016313A (en)*2020-09-082020-12-01迪爱斯信息技术股份有限公司Spoken language element identification method and device and alarm situation analysis system
CN112131880A (en)*2020-09-252020-12-25泰康保险集团股份有限公司Named entity identification method, device and system
CN112434520A (en)*2020-11-112021-03-02北京工业大学Named entity recognition method and device and readable storage medium
CN112507698A (en)*2020-12-072021-03-16深圳市优必选科技股份有限公司Word vector generation method and device, terminal equipment and computer readable storage medium
CN113065339A (en)*2021-04-122021-07-02平安国际智慧城市科技股份有限公司Automatic error correction method, device and equipment for Chinese text and storage medium
CN113065346A (en)*2021-04-022021-07-02国网浙江省电力有限公司信息通信分公司Text entity identification method and related device
CN113139385A (en)*2021-05-122021-07-20北京化工大学Electronic medical record named entity recognition method based on word and pronunciation fusion feature model
CN113284499A (en)*2021-05-242021-08-20湖北亿咖通科技有限公司Voice instruction recognition method and electronic equipment
CN113283241A (en)*2020-02-202021-08-20阿里巴巴集团控股有限公司Text recognition method and device, electronic equipment and computer readable storage medium
CN113393844A (en)*2021-06-242021-09-14大唐融合通信股份有限公司Voice quality inspection method, device and network equipment
CN113408273A (en)*2021-06-302021-09-17北京百度网讯科技有限公司Entity recognition model training and entity recognition method and device
CN113516129A (en)*2021-07-152021-10-19广州云从鼎望科技有限公司Business card recognition method and device, computer readable storage medium and control device
CN113807099A (en)*2021-09-222021-12-17北京百度网讯科技有限公司Entity information identification method, entity information identification device, electronic equipment and storage medium
CN114090773A (en)*2021-11-122022-02-25北京明略软件系统有限公司 Short text classification method, system, computer device and readable storage medium
CN114139541A (en)*2021-11-222022-03-04北京中科闻歌科技股份有限公司Named entity identification method, device, equipment and medium
CN114528841A (en)*2022-02-172022-05-24平安普惠企业管理有限公司Entity identification method and device, electronic equipment and storage medium
CN115146644A (en)*2022-09-012022-10-04北京航空航天大学 A multi-feature fusion named entity recognition method for police texts
EP3961475A4 (en)*2019-04-222023-05-03Boe Technology Group Co., Ltd. METHOD AND DEVICE FOR IDENTIFICATION OF DESIGNATED UNITS
CN116151241A (en)*2023-04-192023-05-23湖南马栏山视频先进技术研究院有限公司Entity identification method and device

Citations (5)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN106354701A (en)*2016-08-302017-01-25腾讯科技(深圳)有限公司Chinese character processing method and device
CN106569998A (en)*2016-10-272017-04-19浙江大学Text named entity recognition method based on Bi-LSTM, CNN and CRF
CN107766327A (en)*2017-10-232018-03-06武汉楚鼎信息技术有限公司The method and system of error correction during a kind of name Entity recognition
US20180121799A1 (en)*2016-11-032018-05-03Salesforce.Com, Inc.Training a Joint Many-Task Neural Network Model using Successive Regularization
CN108268444A (en)*2018-01-102018-07-10南京邮电大学A kind of Chinese word cutting method based on two-way LSTM, CNN and CRF

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN106354701A (en)*2016-08-302017-01-25腾讯科技(深圳)有限公司Chinese character processing method and device
CN106569998A (en)*2016-10-272017-04-19浙江大学Text named entity recognition method based on Bi-LSTM, CNN and CRF
US20180121799A1 (en)*2016-11-032018-05-03Salesforce.Com, Inc.Training a Joint Many-Task Neural Network Model using Successive Regularization
CN107766327A (en)*2017-10-232018-03-06武汉楚鼎信息技术有限公司The method and system of error correction during a kind of name Entity recognition
CN108268444A (en)*2018-01-102018-07-10南京邮电大学A kind of Chinese word cutting method based on two-way LSTM, CNN and CRF

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
JINGKANG WANG 等: "Multiple Character Embeddings for Chinese Word Segmentation", 《HTTPS://ARXIV.ORG》*
买买提阿依甫 等: "基于BiLSTM-CNN-CRF模型的维吾尔文命名实体识别", 《计算机工程》*
司念文 等: "基于注意力长短时记忆网络的中文词性标注模型", 《计算机科学》*

Cited By (77)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN111507109A (en)*2019-01-142020-08-07阿里巴巴集团控股有限公司Named entity identification method and device of electronic medical record
CN111435411A (en)*2019-01-152020-07-21菜鸟智能物流控股有限公司Named body type identification method and device and electronic equipment
CN109871535A (en)*2019-01-162019-06-11四川大学 A French Named Entity Recognition Method Based on Deep Neural Network
CN109858039A (en)*2019-03-012019-06-07北京奇艺世纪科技有限公司A kind of text information identification method and identification device
CN109858039B (en)*2019-03-012023-09-05北京奇艺世纪科技有限公司Text information identification method and identification device
CN109960728A (en)*2019-03-112019-07-02北京市科学技术情报研究所(北京市科学技术信息中心)A kind of open field conferencing information name entity recognition method and system
CN109960728B (en)*2019-03-112021-01-22北京市科学技术情报研究所(北京市科学技术信息中心)Method and system for identifying named entities of open domain conference information
CN111695385A (en)*2019-03-152020-09-22杭州海康威视数字技术股份有限公司Text recognition method, device and equipment
CN111695385B (en)*2019-03-152023-09-26杭州海康威视数字技术股份有限公司Text recognition method, device and equipment
WO2020192237A1 (en)*2019-03-222020-10-01北京京东尚科信息技术有限公司Semantic recognition method, device and system based on artificial intelligence, and storage medium
CN109993227B (en)*2019-03-292021-09-24京东方科技集团股份有限公司 Method, system, device and medium for automatically adding International Classification of Diseases codes
CN109918681B (en)*2019-03-292023-01-31哈尔滨理工大学 A Semantic Matching Method for Fusion Questions Based on Chinese Characters and Pinyin
CN109993227A (en)*2019-03-292019-07-09京东方科技集团股份有限公司 Method, system, device and medium for automatically adding International Classification of Diseases codes
CN109918681A (en)*2019-03-292019-06-21哈尔滨理工大学 A Semantic Matching Method for Fusion Problems Based on Chinese Characters and Pinyin
CN111859963B (en)*2019-04-082024-06-11中移(苏州)软件技术有限公司Named entity recognition method, device, apparatus and computer readable storage medium
CN111859963A (en)*2019-04-082020-10-30中移(苏州)软件技术有限公司 Named entity identification method, apparatus, apparatus and computer-readable storage medium
CN111831818A (en)*2019-04-152020-10-27北京京东尚科信息技术有限公司 Text information identification method, device and storage medium
EP3961475A4 (en)*2019-04-222023-05-03Boe Technology Group Co., Ltd. METHOD AND DEVICE FOR IDENTIFICATION OF DESIGNATED UNITS
CN110069781B (en)*2019-04-242022-11-18北京奇艺世纪科技有限公司Entity label identification method and related equipment
CN110069781A (en)*2019-04-242019-07-30北京奇艺世纪科技有限公司A kind of recognition methods of entity tag and relevant device
CN110298019A (en)*2019-05-202019-10-01平安科技(深圳)有限公司Name entity recognition method, device, equipment and computer readable storage medium
CN110298019B (en)*2019-05-202023-04-18平安科技(深圳)有限公司Named entity recognition method, device, equipment and computer readable storage medium
CN111985235B (en)*2019-05-232024-05-07北京地平线机器人技术研发有限公司Text processing method, text processing device, computer readable storage medium and electronic equipment
CN111985235A (en)*2019-05-232020-11-24北京地平线机器人技术研发有限公司Text processing method and device, computer readable storage medium and electronic equipment
CN110210037B (en)*2019-06-122020-04-07四川大学Syndrome-oriented medical field category detection method
CN110210037A (en)*2019-06-122019-09-06四川大学Category detection method towards evidence-based medicine EBM field
CN110348021A (en)*2019-07-172019-10-18湖北亿咖通科技有限公司Character string identification method, electronic equipment, storage medium based on name physical model
CN110413743B (en)*2019-08-092022-05-06安徽科大讯飞医疗信息技术有限公司Key information extraction method, device, equipment and storage medium
CN110413743A (en)*2019-08-092019-11-05安徽科大讯飞医疗信息技术有限公司A kind of key message abstracting method, device, equipment and storage medium
CN110782002A (en)*2019-09-122020-02-11成都四方伟业软件股份有限公司LSTM neural network training method and device
CN110782002B (en)*2019-09-122022-04-05成都四方伟业软件股份有限公司LSTM neural network training method and device
CN111339764A (en)*2019-09-182020-06-26华为技术有限公司 A Chinese named entity recognition method and device
CN110619877A (en)*2019-09-272019-12-27安徽咪鼠科技有限公司Voice recognition man-machine interaction method, device and system applied to laser pen and storage medium
CN110929749A (en)*2019-10-152020-03-27平安科技(深圳)有限公司Text recognition method, text recognition device, text recognition medium and electronic equipment
CN110929749B (en)*2019-10-152022-04-29平安科技(深圳)有限公司Text recognition method, text recognition device, text recognition medium and electronic equipment
CN110827831A (en)*2019-11-152020-02-21广州洪荒智能科技有限公司Voice information processing method, device, equipment and medium based on man-machine interaction
CN111222335A (en)*2019-11-272020-06-02上海眼控科技股份有限公司Corpus correction method and device, computer equipment and computer-readable storage medium
CN111125386A (en)*2019-12-022020-05-08腾讯科技(北京)有限公司Media resource processing method and device, storage medium and electronic device
CN111104800A (en)*2019-12-242020-05-05东软集团股份有限公司Entity identification method, device, equipment, storage medium and program product
CN111104800B (en)*2019-12-242024-01-23东软集团股份有限公司Entity identification method, entity identification device, entity identification equipment, storage medium and program product
CN111191459A (en)*2019-12-252020-05-22医渡云(北京)技术有限公司Text processing method and device, readable medium and electronic equipment
CN111191459B (en)*2019-12-252023-12-12医渡云(北京)技术有限公司Text processing method and device, readable medium and electronic equipment
CN113283241B (en)*2020-02-202022-04-29阿里巴巴集团控股有限公司Text recognition method and device, electronic equipment and computer readable storage medium
CN113283241A (en)*2020-02-202021-08-20阿里巴巴集团控股有限公司Text recognition method and device, electronic equipment and computer readable storage medium
CN111444719A (en)*2020-03-172020-07-24车智互联(北京)科技有限公司Entity identification method and device and computing equipment
CN111444719B (en)*2020-03-172023-10-20车智互联(北京)科技有限公司Entity identification method and device and computing equipment
CN111680513A (en)*2020-05-292020-09-18平安银行股份有限公司Feature information identification method and device and computer readable storage medium
CN111680513B (en)*2020-05-292024-03-29平安银行股份有限公司Feature information identification method and device and computer readable storage medium
CN111859994A (en)*2020-06-082020-10-30北京百度网讯科技有限公司 Machine translation model acquisition and text translation method, device and storage medium
CN111859994B (en)*2020-06-082024-01-23北京百度网讯科技有限公司Machine translation model acquisition and text translation method, device and storage medium
CN111738808A (en)*2020-07-242020-10-02浙江口碑网络技术有限公司Data processing method, device and equipment
CN111967264A (en)*2020-08-262020-11-20湖北亿咖通科技有限公司Named entity identification method
CN112016313A (en)*2020-09-082020-12-01迪爱斯信息技术股份有限公司Spoken language element identification method and device and alarm situation analysis system
CN112016313B (en)*2020-09-082024-02-13迪爱斯信息技术股份有限公司Spoken language element recognition method and device and warning analysis system
CN112131880B (en)*2020-09-252023-11-07泰康保险集团股份有限公司Named entity identification method, device and system
CN112131880A (en)*2020-09-252020-12-25泰康保险集团股份有限公司Named entity identification method, device and system
CN112434520A (en)*2020-11-112021-03-02北京工业大学Named entity recognition method and device and readable storage medium
CN112507698A (en)*2020-12-072021-03-16深圳市优必选科技股份有限公司Word vector generation method and device, terminal equipment and computer readable storage medium
CN112507698B (en)*2020-12-072024-05-24深圳市优必选科技股份有限公司Word vector generation method, device, terminal equipment and computer readable storage medium
CN113065346A (en)*2021-04-022021-07-02国网浙江省电力有限公司信息通信分公司Text entity identification method and related device
CN113065339A (en)*2021-04-122021-07-02平安国际智慧城市科技股份有限公司Automatic error correction method, device and equipment for Chinese text and storage medium
CN113065339B (en)*2021-04-122023-06-30平安国际智慧城市科技股份有限公司Automatic error correction method, device and equipment for Chinese text and storage medium
CN113139385A (en)*2021-05-122021-07-20北京化工大学Electronic medical record named entity recognition method based on word and pronunciation fusion feature model
CN113139385B (en)*2021-05-122024-05-14北京化工大学Electronic medical record named entity recognition method based on character and word pronunciation fusion feature model
CN113284499A (en)*2021-05-242021-08-20湖北亿咖通科技有限公司Voice instruction recognition method and electronic equipment
CN113393844B (en)*2021-06-242022-12-06大唐融合通信股份有限公司Voice quality inspection method, device and network equipment
CN113393844A (en)*2021-06-242021-09-14大唐融合通信股份有限公司Voice quality inspection method, device and network equipment
CN113408273A (en)*2021-06-302021-09-17北京百度网讯科技有限公司Entity recognition model training and entity recognition method and device
CN113408273B (en)*2021-06-302022-08-23北京百度网讯科技有限公司Training method and device of text entity recognition model and text entity recognition method and device
CN113516129A (en)*2021-07-152021-10-19广州云从鼎望科技有限公司Business card recognition method and device, computer readable storage medium and control device
CN113807099B (en)*2021-09-222024-02-13北京百度网讯科技有限公司Entity information identification method, device, electronic equipment and storage medium
CN113807099A (en)*2021-09-222021-12-17北京百度网讯科技有限公司Entity information identification method, entity information identification device, electronic equipment and storage medium
CN114090773A (en)*2021-11-122022-02-25北京明略软件系统有限公司 Short text classification method, system, computer device and readable storage medium
CN114139541A (en)*2021-11-222022-03-04北京中科闻歌科技股份有限公司Named entity identification method, device, equipment and medium
CN114528841A (en)*2022-02-172022-05-24平安普惠企业管理有限公司Entity identification method and device, electronic equipment and storage medium
CN115146644A (en)*2022-09-012022-10-04北京航空航天大学 A multi-feature fusion named entity recognition method for police texts
CN116151241A (en)*2023-04-192023-05-23湖南马栏山视频先进技术研究院有限公司Entity identification method and device

Similar Documents

PublicationPublication DateTitle
CN109165384A (en)A kind of name entity recognition method and device
CN110390009B (en)Generating topic-based summaries of text content
CN110909548B (en)Chinese named entity recognition method, device and computer readable storage medium
CN108717406B (en)Text emotion analysis method and device and storage medium
CN107729309B (en) A method and device for Chinese semantic analysis based on deep learning
CN109902307B (en)Named entity recognition method, named entity recognition model training method and device
CN109858039B (en)Text information identification method and identification device
CN111310440B (en)Text error correction method, device and system
CN112860919B (en)Data labeling method, device, equipment and storage medium based on generation model
CN111461301B (en)Serialized data processing method and device, and text processing method and device
CN110597994A (en) Event element identification method and device
CN111191428A (en) Comment information processing method, apparatus, computer equipment and medium
CN113626576B (en) Method, device, terminal and storage medium for extracting relational features in remote supervision
CN111666766B (en)Data processing method, device and equipment
CN111414561A (en) Method and apparatus for presenting information
CN110909549A (en) Method, device and storage medium for segmenting ancient Chinese
CN113158656B (en)Ironic content recognition method, ironic content recognition device, electronic device, and storage medium
CN112016314A (en)Medical text understanding method and system based on BERT model
CN108776677B (en)Parallel sentence library creating method and device and computer readable storage medium
CN114328934B (en)Attention mechanism-based multi-label text classification method and system
US20240086452A1 (en)Tracking concepts within content in content management systems and adaptive learning systems
CN113505786A (en)Test question photographing and judging method and device and electronic equipment
CN109582792A (en)A kind of method and device of text classification
CN113360654A (en)Text classification method and device, electronic equipment and readable storage medium
CN112949320A (en)Sequence labeling method, device, equipment and medium based on conditional random field

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination
RJ01Rejection of invention patent application after publication
RJ01Rejection of invention patent application after publication

Application publication date:20190108


[8]ページ先頭

©2009-2025 Movatter.jp