CN114254028A - Event attribute extraction method and device, electronic equipment and storage medium - Google Patents
Event attribute extraction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN114254028A CN114254028A CN202111563341.6A CN202111563341A CN114254028A CN 114254028 A CN114254028 A CN 114254028A CN 202111563341 A CN202111563341 A CN 202111563341A CN 114254028 A CN114254028 A CN 114254028A
- Authority
- CN
- China
- Prior art keywords
- text
- unit
- text unit
- event
- ith
- 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
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 47
- 238000000034 method Methods 0.000 claims description 27
- 238000012545 processing Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 11
- 238000013145 classification model Methods 0.000 claims description 6
- 230000004927 fusion Effects 0.000 claims description 6
- 239000000126 substance Substances 0.000 claims description 2
- 238000013135 deep learning Methods 0.000 abstract description 4
- 238000013473 artificial intelligence Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 9
- 238000004891 communication Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 230000009471 action Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000006399 behavior Effects 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/258—Data format conversion from or to a database
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/38—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/38—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/387—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
Abstract
The disclosure provides an event attribute extraction method and device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence such as knowledge maps and deep learning. The specific implementation scheme is as follows: obtaining an event type corresponding to the target text based on the title of the target text; obtaining suggestive information based on the event type and the title; obtaining the probability of each text unit related to the event attribute based on the suggestive information and each text unit in the target text; and determining the event attribute of the target text based on the probability that each text unit is related to the event attribute.
Description
Technical Field
The disclosure relates to the technical field of artificial intelligence such as knowledge maps and deep learning.
Background
Event attribute extraction refers to analyzing attributes of an event from a text, and a typical application is to analyze an information text so as to analyze various attributes of a subject, an object, time, a place and the like of the information event. The event attribute extraction is an important way for converting rich unstructured texts in the objective world into structured knowledge, so that the method has wide application prospect.
Disclosure of Invention
The disclosure provides an event attribute extraction method, an event attribute extraction device, an electronic device and a storage medium.
According to an aspect of the present disclosure, there is provided an event attribute extraction method, including:
obtaining an event type corresponding to the target text based on the title of the target text;
obtaining suggestive information based on the event type and the title;
obtaining the probability of each text unit related to the event attribute based on the suggestive information and each text unit in the target text;
and determining the event attribute of the target text based on the probability that each text unit is related to the event attribute.
According to another aspect of the present disclosure, there is provided an event attribute extraction apparatus including:
the type identification module is used for determining an event type corresponding to the target text based on the title of the target text;
the prompt acquisition module is used for acquiring prompt information based on the event type and the title;
the probability determining module is used for obtaining the probability of each text unit related to the event attribute based on the suggestive information and each text unit in the target text;
and the attribute determining module is used for determining the event attribute of the target text based on the probability that each text unit is related to the event attribute.
According to another aspect of the present disclosure, there is provided an electronic device including:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform a method according to any one of the embodiments of the present disclosure.
According to another aspect of the present disclosure, there is provided a non-transitory computer readable storage medium having stored thereon computer instructions for causing a computer to perform a method in any of the embodiments of the present disclosure.
According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the method in any of the embodiments of the present disclosure.
According to the technology disclosed by the invention, the title of the target text and the event type determined based on the title are utilized to obtain the suggestive information, and the suggestive information is utilized to assist the analysis of each text unit in the target text, so that the information of various events mixed in the analysis result can be avoided, the correlation between the text and the event attribute can be identified more specifically, and the accuracy of extracting the event attribute is improved.
It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The drawings are included to provide a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:
FIG. 1 is a schematic diagram of an event attribute extraction method according to an embodiment of the present disclosure;
FIG. 2 is a schematic diagram of an event attribute extraction method according to another embodiment of the present disclosure;
FIG. 3 is a schematic diagram of an application example of an event attribute extraction method according to an embodiment of the present disclosure;
FIG. 4 is a schematic diagram of an event attribute extraction apparatus according to an embodiment of the present disclosure;
FIG. 5 is a schematic diagram of an event attribute extraction device according to another embodiment of the present disclosure;
fig. 6 is a block diagram of an electronic device for implementing an event attribute extraction method according to an embodiment of the present disclosure.
Detailed Description
Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
For the convenience of understanding some embodiments of the present disclosure, the following description of the related art is provided, and the following related art may be arbitrarily combined with other technical solutions of the embodiments of the present disclosure as an alternative, and all of them belong to the protection scope of the embodiments of the present disclosure.
In the related art, for example, the generic attribute extraction may be implemented based on a pattern matching manner. For example, the extraction pattern is summarized by the domain expert for a specific domain background, and then the extraction of the event attribute is performed under the guidance of the given extraction pattern. In order to reduce the cost of manually constructing the extraction mode, a method for introducing machine learning to assist in constructing the extraction mode is also researched. The technical scheme seriously depends on the quality of the manually constructed matching pattern, so that the construction is generally carried out by field experts, and the cost is higher. In addition, different extraction modes often require different extraction methods to be constructed, and therefore, the extraction technology is less portable.
By way of example, generic attribute extraction may also be implemented based on machine learning. Wherein the extraction problem of various event attributes is modeled as a plurality of classification problems. With the technical development of deep learning, more and more schemes adopt a deep learning method to realize the extraction process of text features. Although the extraction technology has strong portability, a large amount of marking data needs to be constructed to achieve a good extraction effect. Meanwhile, under the extraction technology, most of the event attributes are directly extracted from sentences/chapters, so that the final extraction result has no pertinence, the extraction result is mixed, and complete structured event knowledge cannot be finally obtained.
The technical solutions of the following embodiments of the present disclosure are mainly used to solve at least one of the above-mentioned problems.
Fig. 1 is a schematic diagram of an event attribute extraction method according to an embodiment of the present disclosure. As shown in fig. 1, the method may include:
step S110, obtaining an event type corresponding to the target text based on the title of the target text;
step S120, obtaining suggestive information based on the event type and the title;
step S130, based on the suggestive information and each text unit in the target text, obtaining the probability of each text unit related to the event attribute;
step S140, determining the event attribute of the target text based on the probability that each text unit is related to the event attribute.
Illustratively, the target text may include text of the attributes of the event to be extracted, such as various announcements, informational text, research reports, and the like. In some scenarios, the event type may be a business segment type of the target text, including, for example, financial events, scientific events, sporting events, and the like. In other scenarios, the event type may be event dynamics information for the target text, including, for example, loss events, profit events, reward events, and the like.
For example, in step S110, the event type of the target text may be obtained by text matching based on the title of the target text and a plurality of preset event types. The deep learning model, such as a classifier, may also be used to analyze the title of the target text for classification processing, so as to obtain the event type of the target text.
It should be noted that one target text may correspond to multiple event types, for example, one information event title may describe events under multiple types. In the case where the target text corresponds to a plurality of event types, suggestive information can be obtained based on the plurality of event types and the title.
Illustratively, one or more event types and titles may be concatenated, for example, with a special linker to link the event types and titles to obtain the suggestive information. It can be understood that the suggestive information includes type information and title information of the event related to the target text, so that the suggestive information can embody main event information in the target text to assist the extraction of the event attribute, and a plurality of event mashups can be avoided.
For example, in the embodiment of the present disclosure, the event attribute may include a subject, an object, time, place, and the like of the event. Wherein the subject of the event is the issuer of the action, behavior or activity. The object of an event is the recipient of an action, behavior or activity. The time of the event is time information such as the occurrence time and the end time of the action, the behavior or the activity. The location of the event is the place of occurrence of the action, activity or activity, the place of publication, etc.
For example, in the embodiments of the present disclosure, a text unit may refer to elements constituting a text, such as information of a word, and the like. The text unit is related to the event attribute, and may refer to that the text unit is an attribute value of the event attribute, and may also refer to that the text unit is positioning information of the event attribute, where the positioning information may include a start position and/or an end position of the event attribute. For example, if a text element is a word, then a certain text element (a certain word) is associated with the body of the event, which may refer to: the word is the beginning or ending position of the body. If the text unit is a word, a certain text unit (a certain word) is related to the main body of the event, which may refer to: the word is the subject of the event.
In practical application, the probability of each text unit and each event attribute can be obtained for a plurality of event attributes respectively. Thus, a plurality of event attributes of the target text may be determined based on the probability that each text unit is associated with each event attribute.
For example, in step S140, at least one text unit having a probability related to the event attribute meeting a preset condition (for example, the probability is the maximum) may be determined in the target text based on the probability related to the event attribute of each text, and the event attribute of the target text may be obtained based on the determined at least one text unit.
For example, for each word in the target text, a probability is obtained that it is the starting or ending position of the subject of the event. Then, in combination with the probabilities of all the words in the target text, a word having the highest probability of the start position as the subject of the event and a word having the highest probability of the end position as the subject of the event are determined, and the text between the start position and the end position is determined as the subject of the event.
According to the method, the suggestive information is obtained by utilizing the title of the target text and the event type determined based on the title, and the suggestive information is utilized to assist the analysis of each text unit in the target text, so that the information of various events mixed in the analysis result can be avoided, the correlation between the text and the event attribute can be identified more specifically, and the accuracy of extracting the event attribute is improved.
In an exemplary embodiment, in step S110, obtaining an event type corresponding to the target text based on the title of the target text may include:
and classifying the title of the target text by using the event type classification model to obtain the event type corresponding to the target text.
The event type classification model may be a pre-trained neural network model, and may be a multi-classifier, for example.
According to the exemplary embodiment, the accuracy of the obtained event type can be improved, so that the suggestive information is obtained based on the event type with high accuracy, the event attribute extraction is assisted, and the accuracy of the event attribute extraction can be further improved.
In an exemplary embodiment, as shown in fig. 2, the step S130 of obtaining, based on the suggestive information and each text unit in the target text, a probability that each text unit is related to the event attribute may include:
step S210, coding is carried out based on the suggestive information and the ith text unit in the target text, and the characteristic representation of the ith text unit is obtained; i is an integer of 1 or more;
step S220, classifying the ith text unit based on the feature representation of the ith text unit to obtain the probability of the ith text unit related to the event attribute.
Wherein, the ith text unit can be any text unit in the target text. That is, for any text unit in the target text, the exemplary embodiment can be used to obtain the probability that it is related to the event attribute.
According to the exemplary embodiment, the text units are encoded based on the suggestive information, so that the feature representation of each text unit is determined based on the event type and the title information, the pertinence and the accuracy of the feature representation are improved, the obtained probability is relatively more accurate when the classification processing is performed based on the feature representation, and the accuracy of the event attribute extraction is further improved.
In an exemplary embodiment, the step S210, encoding based on the suggestive information and the ith text unit in the target text to obtain the feature representation of the ith text unit, includes:
and performing weighted fusion on the suggestive information, the ith text unit and the context information of the ith text unit by using an encoder to obtain the feature representation of the ith text unit.
That is, a text unit may be encoded using an encoder; also, the encoding process takes into account not only the suggestive information and the text units themselves, but also the context information of the text units.
For example, the context information of the ith text unit may include the first n text units of the ith text unit and the last n text units of the ith text unit, where n is an integer greater than or equal to 1.
For example, all the text units in the suggestive information and the target text may be input to the encoder, and any one of the text units is interacted with the suggestive information and the context information in the encoder, and the specific interaction process may be to set different weights for the text unit itself, the suggestive information, and the context information, and perform weighting fusion. And obtaining the feature representation of the text unit based on the interacted information, and then carrying out classification processing to obtain the probability of the text unit related to the event attribute.
According to the exemplary embodiment, the context information of the ith text unit is introduced and is subjected to weighted fusion with the ith text and the suggestive information, the pertinence and the accuracy of feature representation can be improved, the probability obtained by classification processing based on the feature representation is relatively more accurate, and the accuracy of event attribute extraction is further improved.
In an exemplary embodiment, in step S220, classifying the ith text unit based on the feature representation of the ith text unit to obtain a probability that the ith text unit is related to the event attribute may include:
classifying and judging whether the ith text unit is the positioning information of the event attribute by utilizing a multilayer pointer network to obtain the probability that the ith text unit is the positioning information of the event attribute;
wherein the positioning information comprises a start position and/or an end position of the event attribute.
For example, the probability that each text unit is the starting position and/or the ending position of an attribute value under a certain attribute can be predicted by utilizing each layer of pointer network in the multilayer pointer network. For the feature representation of each text unit, inputting it into the multi-layer pointer network, the probability that each text unit is used as the start position and/or the end position can be obtained. After decoding, the attribute values under each event attribute, that is, the event attributes in the target text, can be obtained.
According to the exemplary embodiment, the positioning information of whether the ith text unit is the event attribute is classified and judged by using the multilayer pointer network, so that the accuracy of the probability can be improved, and the accuracy of the event attribute extraction can be further improved.
A specific application example of the event attribute extraction method in the embodiment of the present disclosure is given below.
In the application example, for a given information text, firstly, a pre-trained event type classification model is adopted to classify the title of the information text, so as to obtain the event type of the corresponding information event. Here, one information text may correspond to a plurality of predefined event types, and represents that the information text describes events under a plurality of types.
As shown in FIG. 3, the event type and the title of the information text can be linked by a special link character as the suggestive information (Q in the figure)1To QN) And the event attribute is used for indicating the event to which the event attribute to be extracted belongs. At the same time, the suggestive information and the information text (including the text unit T) are combined1To TN) Splicing with special link symbol, sending into encoder, and interacting the suggestive information with the information text to obtain the feature representation (such as H in the figure) of each text unit in the information text1To HN). Classifying each text unit (or element) in the information text by using a multilayer pointer network, and predicting the probability of the text unit (or element) serving as the starting position and the ending position of each general attribute; finally, all the attribute values of the information text are obtained by decoding.
It can be seen that according to the method, the suggestive information is obtained by using the title of the target text and the event type determined based on the title, and the suggestive information is used for assisting the analysis of each text unit in the target text, so that the information of various events mixed in the analysis result can be avoided, the correlation between the text and the event attribute can be identified more specifically, and the accuracy of extracting the event attribute is improved.
It should be noted that, in the technical solution of the present disclosure, the acquisition, storage, application, and the like of the personal information of the related user all conform to the regulations of the relevant laws and regulations, and do not violate the good custom of the public order.
As an implementation of the foregoing methods, an embodiment of the present disclosure further provides an event attribute extraction device, as shown in fig. 4, where the device includes:
the type identification module 410 is used for determining an event type corresponding to the target text based on the title of the target text;
a prompt obtaining module 420, configured to obtain prompt information based on the event type and the title;
the probability determination module 430 is used for obtaining the probability of each text unit related to the event attribute based on the suggestive information and each text unit in the target text;
and an attribute determining module 440, configured to determine an event attribute of the target text based on the probability that each text unit is associated with the event attribute.
Illustratively, as shown in fig. 5, the probability determination module 430 includes:
the encoding unit 510 is configured to encode based on the suggestive information and an ith text unit in the target text, so as to obtain a feature representation of the ith text unit; i is an integer of 1 or more;
the classifying unit 520 is configured to perform classification processing based on the feature representation of the ith text unit to obtain a probability that the ith text unit is the positioning information of the event attribute.
Exemplarily, the encoding unit 510 is specifically configured to:
and performing weighted fusion on the suggestive information, the ith text unit and the context information of the ith text unit by using an encoder to obtain the feature representation of the ith text unit.
Exemplarily, the classification unit 520 is specifically configured to:
and classifying and judging whether the ith text unit is the positioning information of the event attribute by utilizing a multilayer pointer network to obtain the probability that the ith text unit is the positioning information of the event attribute.
Illustratively, the type identification module 410 is specifically configured to:
and classifying the title of the target text by using the event type classification model, and determining the event type corresponding to the target text.
The functions of each unit, module or sub-module in each apparatus in the embodiments of the present disclosure may refer to the corresponding description in the above method embodiments, and are not described herein again.
The present disclosure also provides an electronic device, a readable storage medium, and a computer program product according to embodiments of the present disclosure.
FIG. 6 illustrates a schematic block diagram of an example electronic device 600 that can be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.
As shown in fig. 6, the electronic device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM)602 or a computer program loaded from a storage unit 608 into a Random Access Memory (RAM) 603. In the RAM 603, various programs and data necessary for the operation of the electronic apparatus 600 can also be stored. The calculation unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.
Various components in the electronic device 600 are connected to the I/O interface 605, including: an input unit 606 such as a keyboard, a mouse, or the like; an output unit 607 such as various types of displays, speakers, and the like; a storage unit 608, such as a magnetic disk, optical disk, or the like; and a communication unit 609 such as a network card, modem, wireless communication transceiver, etc. The communication unit 609 allows the electronic device 600 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunication networks.
The computing unit 601 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various dedicated Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and so forth. The calculation unit 601 executes the respective methods and processes described above, such as the event attribute extraction method. For example, in some embodiments, the event attribute extraction method may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 600 via the ROM 602 and/or the communication unit 609. When a computer program is loaded into RAM 603 and executed by the computing unit 601, one or more steps of the event property extraction method described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to perform the event attribute extraction method in any other suitable manner (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.
The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server with a combined blockchain.
It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be executed in parallel or sequentially or in different orders, and are not limited herein as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved.
The above detailed description should not be construed as limiting the scope of the disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present disclosure should be included in the scope of protection of the present disclosure.
Claims (13)
1. An event attribute extraction method comprises the following steps:
obtaining an event type corresponding to a target text based on a title of the target text;
obtaining suggestive information based on the event type and the title;
obtaining the probability of each text unit related to the event attribute based on the suggestive information and each text unit in the target text;
and determining the event attribute of the target text based on the probability that each text unit is related to the event attribute.
2. The method of claim 1, wherein said deriving a probability that each text unit is associated with an event attribute based on the suggestive information and each text unit in the target text comprises:
coding based on the suggestive information and the ith text unit in the target text to obtain the feature representation of the ith text unit; i is an integer of 1 or more;
and classifying the ith text unit based on the feature representation of the ith text unit to obtain the probability of the ith text unit being related to the event attribute.
3. The method of claim 2, wherein the encoding based on the suggestive information and an ith text unit in the target text to obtain a feature representation of the ith text unit comprises:
and carrying out weighted fusion on the suggestive information, the ith text unit and the context information of the ith text unit by utilizing an encoder to obtain the feature representation of the ith text unit.
4. The method according to claim 2 or 3, wherein the classifying the ith text unit based on the feature representation of the ith text unit to obtain the probability that the ith text unit is related to the event attribute comprises:
classifying and judging whether the ith text unit is the positioning information of the event attribute by utilizing a multilayer pointer network to obtain the probability that the ith text unit is the positioning information of the event attribute;
wherein the positioning information comprises a start position and/or an end position of the event attribute.
5. The method according to any one of claims 1-4, wherein the obtaining an event type corresponding to the target text based on a title of the target text comprises:
and classifying the title of the target text by using an event type classification model to obtain an event type corresponding to the target text.
6. An event attribute extraction device, comprising:
the type identification module is used for determining an event type corresponding to a target text based on a title of the target text;
the prompt acquisition module is used for acquiring prompt information based on the event type and the title;
the probability determining module is used for obtaining the probability of each text unit related to the event attribute based on the suggestive information and each text unit in the target text;
and the attribute determining module is used for determining the event attribute of the target text based on the probability that each text unit is related to the event attribute.
7. The apparatus of claim 6, wherein the probability determination module comprises:
the coding unit is used for coding based on the suggestive information and the ith text unit in the target text to obtain the feature representation of the ith text unit; i is an integer of 1 or more;
and the classification unit is used for performing classification processing based on the feature representation of the ith text unit to obtain the probability that the ith text unit is the positioning information of the event attribute.
8. The apparatus according to claim 7, wherein the encoding unit is specifically configured to:
and carrying out weighted fusion on the suggestive information, the ith text unit and the context information of the ith text unit by utilizing an encoder to obtain the feature representation of the ith text unit.
9. The apparatus according to claim 7 or 8, wherein the classification unit is specifically configured to:
and classifying and judging whether the ith text unit is the positioning information of the event attribute by utilizing a multilayer pointer network to obtain the probability that the ith text unit is the positioning information of the event attribute.
10. The apparatus according to any one of claims 6-9, wherein the type identification module is specifically configured to:
and classifying the title of the target text by using an event type classification model, and determining the event type corresponding to the target text.
11. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-5.
12. A non-transitory computer readable storage medium having stored thereon computer instructions for causing a computer to perform the method of any one of claims 1-5.
13. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111563341.6A CN114254028A (en) | 2021-12-20 | 2021-12-20 | Event attribute extraction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111563341.6A CN114254028A (en) | 2021-12-20 | 2021-12-20 | Event attribute extraction method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114254028A true CN114254028A (en) | 2022-03-29 |
Family
ID=80795991
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111563341.6A Pending CN114254028A (en) | 2021-12-20 | 2021-12-20 | Event attribute extraction method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114254028A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9652530B1 (en) * | 2014-08-27 | 2017-05-16 | Google Inc. | Generating and applying event data extraction templates |
| CN111723569A (en) * | 2020-05-21 | 2020-09-29 | 上海明略人工智能(集团)有限公司 | Event extraction method and device and computer readable storage medium |
| CN111966890A (en) * | 2020-06-30 | 2020-11-20 | 北京百度网讯科技有限公司 | Text-based event pushing method and device, electronic equipment and storage medium |
| CN112650919A (en) * | 2020-11-30 | 2021-04-13 | 北京百度网讯科技有限公司 | Entity information analysis method, apparatus, device and storage medium |
| US20210200947A1 (en) * | 2020-03-20 | 2021-07-01 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Event argument extraction method and apparatus and electronic device |
| CN113536784A (en) * | 2021-01-05 | 2021-10-22 | 腾讯科技（深圳）有限公司 | Text processing method and device, computer equipment and storage medium |
-
2021
- 2021-12-20 CN CN202111563341.6A patent/CN114254028A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9652530B1 (en) * | 2014-08-27 | 2017-05-16 | Google Inc. | Generating and applying event data extraction templates |
| US20210200947A1 (en) * | 2020-03-20 | 2021-07-01 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Event argument extraction method and apparatus and electronic device |
| CN111723569A (en) * | 2020-05-21 | 2020-09-29 | 上海明略人工智能(集团)有限公司 | Event extraction method and device and computer readable storage medium |
| CN111966890A (en) * | 2020-06-30 | 2020-11-20 | 北京百度网讯科技有限公司 | Text-based event pushing method and device, electronic equipment and storage medium |
| CN112650919A (en) * | 2020-11-30 | 2021-04-13 | 北京百度网讯科技有限公司 | Entity information analysis method, apparatus, device and storage medium |
| CN113536784A (en) * | 2021-01-05 | 2021-10-22 | 腾讯科技（深圳）有限公司 | Text processing method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112560496A (en) | Training method and device of semantic analysis model, electronic equipment and storage medium | |
| CN113129870B (en) | Training method, device, equipment and storage medium of speech recognition model | |
| CN113190702B (en) | Method and device for generating information | |
| CN114724168A (en) | Training method of deep learning model, text recognition method, text recognition device and text recognition equipment | |
| CN112559885A (en) | Method and device for determining training model of map interest point and electronic equipment | |
| CN113836925A (en) | Training method and device for pre-training language model, electronic equipment and storage medium | |
| CN115099233A (en) | Semantic analysis model construction method and device, electronic equipment and storage medium | |
| CN113904943A (en) | Account detection method and device, electronic equipment and storage medium | |
| CN116578925B (en) | Behavior prediction method, device and storage medium based on feature images | |
| CN113468857A (en) | Method and device for training style conversion model, electronic equipment and storage medium | |
| CN114218940B (en) | Text information processing and model training method, device, equipment and storage medium | |
| CN113360672B (en) | Method, apparatus, device, medium and product for generating knowledge graph | |
| CN114970540A (en) | Method and device for training text audit model | |
| CN114897183A (en) | Problem data processing method, and deep learning model training method and device | |
| CN114254028A (en) | Event attribute extraction method and device, electronic equipment and storage medium | |
| CN114078274A (en) | Face image detection method and device, electronic equipment and storage medium | |
| CN113806541A (en) | Emotion classification method and emotion classification model training method and device | |
| CN114141236A (en) | Language model updating method and device, electronic equipment and storage medium | |
| CN113807391A (en) | Task model training method and device, electronic equipment and storage medium | |
| CN113204616A (en) | Method and device for training text extraction model and extracting text | |
| CN112947928A (en) | Code evaluation method and device, electronic equipment and storage medium | |
| CN115131709B (en) | Video category prediction method, training method and device for video category prediction model | |
| CN115965018B (en) | Training method of information generation model, information generation method and device | |
| CN113361712B (en) | Training method of feature determination model, semantic analysis method, semantic analysis device and electronic equipment | |
| CN113886543A (en) | Method, apparatus, medium, and program product for generating an intent recognition model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220329 |