CN113792145A - Method and device for determining object attribute parameters, terminal equipment and storage medium - Google Patents
Method and device for determining object attribute parameters, terminal equipment and storage medium Download PDFInfo
- Publication number
- CN113792145A CN113792145A CN202111088965.7A CN202111088965A CN113792145A CN 113792145 A CN113792145 A CN 113792145A CN 202111088965 A CN202111088965 A CN 202111088965A CN 113792145 A CN113792145 A CN 113792145A
- Authority
- CN
- China
- Prior art keywords
- target
- comment
- text
- emotion
- texts
- 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
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000003860 storage Methods 0.000 title claims abstract description 16
- 230000008451 emotion Effects 0.000 claims abstract description 159
- 238000012545 processing Methods 0.000 claims abstract description 28
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims description 59
- 238000013145 classification model Methods 0.000 claims description 49
- 238000004590 computer program Methods 0.000 claims description 21
- 238000004891 communication Methods 0.000 claims description 10
- 238000005457 optimization Methods 0.000 claims description 4
- 230000014759 maintenance of location Effects 0.000 claims description 2
- 238000013473 artificial intelligence Methods 0.000 abstract description 7
- 230000002996 emotional effect Effects 0.000 description 14
- 230000008569 process Effects 0.000 description 13
- 238000012552 review Methods 0.000 description 13
- 238000012827 research and development Methods 0.000 description 12
- 238000005516 engineering process Methods 0.000 description 8
- 239000013598 vector Substances 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 230000004044 response Effects 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 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
- G06F16/30—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
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
- G06F40/186—Templates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/263—Language identification
Abstract
The application provides a method, a device, a terminal device and a storage medium for determining object attribute parameters, and relates to the technical field of artificial intelligence. The method comprises the following steps: if the terminal equipment is detected to be in the target running state and an object selection instruction is received, a target comment text for the target object is obtained according to the analysis result of the object selection instruction, so that when the terminal equipment is in the target running state and the object selection instruction is received, the initial condition of the target object can be known quickly through the target comment text, emotion classification processing is carried out on the target comment text, the emotion classification result of the target comment text is obtained, the emotion tendency of a commentator on the target object can be known, the specified attribute parameters of the target object can be further determined according to the emotion classification result, the condition of the target object can be known based on the specified attribute parameters, and the condition of the target object can be better obtained and displayed through the scheme provided by the application.
Description
Technical Field
The present application belongs to the technical field of artificial intelligence, and in particular, to a method, an apparatus, a terminal device, and a storage medium for determining an object attribute parameter.
Background
With the rapid development of the internet, more and more data information is spread on the network. In order to understand a target object, it is often necessary to acquire information associated with the target object in order to understand the target object based on the acquired information. For example, in order to make a decision better, an enterprise manager often needs to know the situation of the enterprise or a competitive enterprise so as to provide a basis for making the decision.
Therefore, a method capable of presenting the situation of the target object is needed.
Disclosure of Invention
The embodiment of the application provides a method, a device, a terminal device and a storage medium for determining object attribute parameters, so as to solve the problem of how to acquire and display the condition of a target object.
In a first aspect, an embodiment of the present application provides a method for determining an object attribute parameter, including:
if the terminal equipment is detected to be in a target running state and an object selection instruction is received, acquiring a target comment text aiming at a target object according to an analysis result of the object selection instruction;
performing emotion classification processing on the target comment text to obtain an emotion classification result of the target comment text;
and determining the designated attribute parameters of the target object according to the emotion types of the target comment texts described in the emotion classification result.
According to the method for determining the object attribute parameters, if it is detected that the terminal device is in the target operation state and receives the object selection instruction, the target comment text for the target object is obtained according to the analysis result of the object selection instruction, so that when the terminal device is in the target operation state and receives the object selection instruction, the preliminary condition of the target object can be quickly known through the target comment text, then emotion classification processing is performed on the target comment text, the emotion classification result of the target comment text is obtained, the emotion tendency of a commentator on the target object is known, the designated attribute parameters of the target object are determined according to the emotion classification result, and the condition of the target object is known based on the designated attribute parameters.
In a second aspect, an embodiment of the present application provides an apparatus for determining an object attribute parameter, including:
the acquisition module is used for acquiring a target comment text aiming at a target object according to an analysis result of an object selection instruction if the terminal equipment is detected to be in a target running state and the object selection instruction is received;
the processing module is used for executing emotion classification processing on the target comment text to obtain an emotion classification result of the target comment text;
and the determining module is used for determining the designated attribute parameters of the target object according to the emotion types of the target comment texts described in the emotion classification result.
In a third aspect, an embodiment of the present application provides a terminal device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the method when executing the computer program.
In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the computer program implements the method.
In a fifth aspect, the present application provides a computer program product, which when run on a terminal device, causes the terminal device to execute the method of any one of the above first aspects.
It is understood that the beneficial effects of the second aspect to the fifth aspect can be referred to the related description of the first aspect, and are not described herein again.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
Fig. 1 is a schematic flowchart of a method for determining an object attribute parameter according to an embodiment of the present application.
Fig. 2 is a flowchart illustrating a method for determining an object attribute parameter according to another embodiment of the present application.
Fig. 3 is a schematic structural diagram of an apparatus for determining an object attribute parameter according to an embodiment of the present application.
Fig. 4 is a schematic structural diagram of a terminal device according to an embodiment of the present application.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details.
As used in this specification and the appended claims, the term "if" may be interpreted in context to mean "when", "upon" or "in response to a determination" or "in response to a detection". Similarly, the phrase "if it is determined" or "if a [ described condition or event ] is detected" may be interpreted contextually to mean "upon determining" or "in response to determining" or "upon detecting [ described condition or event ]" or "in response to detecting [ described condition or event ]".
Furthermore, in the description of the present application and the appended claims, the terms "first," "second," "third," and the like are used for distinguishing between descriptions and not necessarily for describing or implying relative importance.
Reference throughout this specification to "one embodiment" or "some embodiments," or the like, means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," or the like, in various places throughout this specification are not necessarily all referring to the same embodiment, but rather "one or more but not all embodiments" unless specifically stated otherwise. The terms "comprising," "including," "having," and variations thereof mean "including, but not limited to," unless expressly specified otherwise.
In order to explain the technical solution described in the present application, the following description will be given by way of specific examples.
Referring to fig. 1, fig. 1 is a flowchart illustrating an implementation of a method for determining object attribute parameters according to an embodiment of the present disclosure. In this embodiment, the method for determining the object attribute parameters may be applied to terminal devices such as a mobile phone, a tablet computer, a notebook computer, an ultra-mobile personal computer (UMPC), a netbook, and a server.
As shown in fig. 1, a method for determining an object attribute parameter provided in an embodiment of the present application includes the following steps:
s11: and if the terminal equipment is detected to be in the target running state and the object selection instruction is received, acquiring a target comment text aiming at the target object according to the analysis result of the object selection instruction.
As an example of the present application, a terminal device refers to a device for determining specified attribute parameters of a target object. Such as a server or a laptop.
The target operation state refers to an operation state when a specified attribute parameter of a target object can be determined based on a terminal device. For example, after the terminal device is turned on, the operating voltage of the terminal device is in a stable state, so that the terminal device can stably operate, and the state of the corresponding terminal device is established when the software environment for determining the designated attribute parameters of the target object is established.
It is to be understood that the target operating state includes one or more of a physical operating state of the terminal device and/or a state of the software operating environment.
The object selection instruction is an instruction generated by determining an object requiring determination of a specified attribute parameter according to a user requirement. For example, since there are currently a business a and a business B, and a user needs to know about the business B, that is, to specify attribute parameters of the business B, an object selection instruction is generated according to a selection operation of the user.
It is to be understood that at least the information of the target object for which the specified attribute parameter is to be determined is included in the object selection instruction. For example, information capable of identifying the target object, such as the name, number, etc. of the target object, is included.
The target object refers to an object that the user desires to know. For example, an enterprise manager needs to know about the local enterprise or a competing enterprise when it needs to make a decision.
The target comment text refers to text including at least a comment sentence. For example, a target comment text containing a comment sentence "this company has a large sales amount" or a target comment text containing a comment sentence "this company has a large amount of development and innovation products" is produced.
In this embodiment, in order to be able to efficiently determine the object attribute parameters, it is necessary to detect whether the terminal device is in the target operation state and receive the object selection instruction in real time. In addition, since the emotional tendency information of the reviewer on the target object is recorded in the target comment text, in order to determine the attraction of the target object to the reviewer or compare the target object with other objects based on the angle of the reviewer, when the terminal device is detected to be in the target operation state and an object selection instruction is received, the target comment text for the target object is acquired, so that the competitiveness of the target object relative to other objects can be determined subsequently based on the emotional tendency information contained in the target comment text.
For example, in a specific application scenario, in order to determine the competitiveness of a business in the same industry, a target comment text for the business is obtained first, so that the opinion of an external person for the business is known through the target comment text, for example, the evaluation of a reviewer on the aspects of capital activities, market competition, production and management, financial conditions, research and development innovations, and the like of the business is known through the target comment text, and thus the competitiveness of the business in the market is indirectly obtained.
In an embodiment of the present application, in a possible scenario, the comment information for the target object is generally recorded in a text that takes the target object as a mainly described object, and the target comment text is only a part of the text, so a specific implementation manner for obtaining the target comment text for the target object may include:
a text paragraph is obtained that describes the target object.
Target comment text is extracted from the text passage.
As an example of the present application, a text paragraph refers to a part of text that mainly describes an object with a target object. For example, a second paragraph of an article describing a production business of a business includes a target review text for the business.
In this embodiment, after a text paragraph used for describing a target object is acquired, text information in the text paragraph is identified, and a target comment text is extracted according to an identification result.
It is understood that there may be one or more text paragraphs used to describe the target object.
In some embodiments, in order to quickly acquire a target comment text, after a text paragraph describing a target object is acquired, punctuation information in the text paragraph is identified, and a position corresponding to a target punctuation in the text paragraph is determined from the punctuation information; and identifying upper and lower text information of the position corresponding to the target punctuation mark in the text paragraph according to the position corresponding to the target punctuation mark, and extracting to obtain a target comment text according to the identification unlocking.
For example, for the text paragraph "this enterprise went up in 2004, the number of employees was 1200. In 2020, the research and development investment of the enterprise accounts for 10% and the net profit is 10 billion yuan. Therefore, the enterprise pays more attention to the investment of research and development and innovation and has better profit capacity! "wherein the target punctuation in the text passage is"! And the position of the target punctuation mark is the end of the text paragraph, so that the upper and lower text information of the position corresponding to the target punctuation mark in the text paragraph is recognized, namely, the comment sentence is recognized, the enterprise pays more attention to the investment of research and development and innovation, and has better profitability, namely, the comment sentence is used as the target comment text.
In an embodiment, since grammars corresponding to different languages are different, after a text paragraph describing a target object is obtained, type information of the language of the text paragraph is identified, a target comment text extraction template is determined according to the type information of the identified language, and text information recorded in the text paragraph is identified according to the target comment text extraction template. And when the comment sentences corresponding to the target comment text extraction template are identified, the identified comment sentences are used as target comment texts.
In the present embodiment, a target comment text extraction template is generated based on the type information and grammar habit information of the language.
In one embodiment, in order to better and more comprehensively understand the target object, a target comment text for the target object within a past preset time period is acquired. The preset time period can be preset according to actual requirements. For example, the past half year time calculated at the current time point.
S12: and performing emotion classification processing on the target comment text to obtain an emotion classification result of the target comment text.
As an example of the application, the emotion classification result is used for describing emotion tendency classification corresponding to comment information contained in the target comment text. For example, the emotional propensity classification may include one or more of positive, neutral, or negative.
In this embodiment, in order to better understand the situation of the target object, after the target comment text for the target object is obtained, emotion classification processing is performed on the target comment text to determine emotion tendency classification corresponding to the target comment text, and an emotion classification result is obtained, so that emotion tendency of a reviewer when commenting on the target object can be known through the emotion classification result.
When emotion classification processing is executed on the target comment text, information used for describing emotional tendency and contained in the target comment text is identified, and an emotion classification result of the target comment text is generated according to the information used for describing emotional tendency and contained in the identified target comment text.
For example, in a specific application scenario, the target comment text is a text that is commented on a certain aspect of an enterprise, so in order to know the emotional tendency of a reviewer when commenting on an enterprise, the target comment text is subjected to emotion classification processing to determine the emotional tendency of the reviewer when commenting on the enterprise, for example, the emotional tendency of an analyst when commenting on the research and development investment of the enterprise, such as a high research and development investment.
It can be understood that, in the application, when a plurality of target comment texts for a target object are acquired, the emotion classification result includes an emotion classification corresponding to each of the plurality of target comment texts.
In some embodiments, performing emotion classification processing on the target comment text to obtain an emotion classification result of the target comment text includes: and determining the comment time of the comment text, if the comment time of the comment text falls into a preset time period, taking the comment text as a target comment text, and performing emotion classification processing on the target comment text to obtain an emotion classification result of the target comment text.
As a possible implementation manner of the present application, in order to conveniently obtain emotion tendency information of a target comment text for a target object, an emotion classification process is performed on the target comment text, and a specific implementation manner of obtaining an emotion classification result of the target comment text includes:
and inputting the target comment text into the trained text sentiment classification model for processing to obtain a sentiment classification result of the target comment text.
The text emotion classification model is obtained by training in the following mode:
and acquiring sample comment data marked with emotion categories.
And training to obtain a text emotion classification model by taking the sample comment data as a training set.
In the present embodiment, the emotion classification refers to an emotion classification corresponding to emotion tendency information included in comment data. For example, if the comment data is "the financial liability ratio of the company is high" and the emotion type of the comment data is labeled, the corresponding emotion type is "positive".
It can be understood that, in order to improve the efficiency of data processing, the target comment text is input into the trained text sentiment classification model for processing, so that the target comment text is processed through the trained text sentiment classification model to obtain a sentiment classification result of the target comment text.
In order to facilitate the text emotion classification model to accurately process the target comment text to obtain an emotion classification result of the target comment text, sample comment data with emotion classes labeled are used as a training set, and the text emotion classification model is obtained through training.
In one embodiment, the process of inputting the target comment text into the trained text emotion classification model for processing comprises the following steps: the method comprises the steps of coding a target comment text through a trained text sentiment classification model to obtain a characteristic vector corresponding to the target comment text, extracting and obtaining head and tail position vectors of attribute words describing sentiment tendency, meanwhile, calculating an average value of the head and tail position vectors of the attribute words, adding the average value with word vectors corresponding to the attribute words in the characteristic vector corresponding to the target comment text to obtain a tensor of the attribute words, so that the average value of the head and tail position vectors of the attribute words is related with the characteristic vector of the target comment text, when a plurality of attribute words exist in the target comment text, the attribute words can be well distinguished, and therefore sentiment classification results corresponding to the target comment text can be obtained through classification based on the tensor of the attribute words.
It can be understood that when the target comment text includes a plurality of attribute words, in order to better optimize the emotion types corresponding to the target comment text, the sub-emotion types of each attribute word are determined first, then the number of the attribute words corresponding to each sub-emotion type is counted, and the emotion types of the target comment text are determined according to the number of the attribute words corresponding to each sub-emotion type.
For example, in a specific scenario, a target comment text is "the research and development investment of the enterprise is large, the research and development yield is also large, and the conversion application based on research and development is few", the attribute words included in the target comment text are "large", "many" and "few", and the sub-emotion categories respectively corresponding to the attribute words are positive, positive and negative, where the number of attribute words corresponding to the positive is 2, and the number of attribute words corresponding to the negative is 1, then the emotion category of the target comment text is positive.
As another possible implementation manner of this embodiment, the training set includes a first sub-training data set and a second sub-training data set, where the first sub-training data set includes class sample comment data labeled with a first emotion category, and the second sub-training data set includes one or more of sample comment data labeled with the first emotion category or sample comment data labeled with a second emotion category.
Taking the sample comment data as a training set, and training to obtain a text emotion classification model, wherein the training comprises the following steps:
and optimizing and updating the parameters of the pre-constructed initial emotion classification model based on the first sub-training data set and the second sub-training data set, and determining the initial emotion classification model after parameter optimization and updating as a text emotion classification model.
In this embodiment, in order to enable the text sentiment classification model to better process the target comment text so as to identify sentiment tendency information contained in the target comment text, parameters of the pre-constructed initial sentiment classification model are optimized and updated based on the first sub-training data set and the second sub-training data set containing sample comment data labeled with different sentiment categories, so that the initial sentiment classification model after parameter optimization and update is determined as the text sentiment classification model.
It can be understood that, in the application, in order to better train the initial emotion classification model by using the second sub-training data set, if the second sub-training data includes the sample comment data labeled with the first emotion category and the sample comment data labeled with the second emotion category, the sample comment data labeled with the first emotion category and the sample comment data labeled with the second emotion category are sorted according to a uniform format.
In one embodiment, a pre-constructed initial emotion classification model is trained based on a first sub-training data set.
And training the pre-constructed initial emotion classification model which is trained and completed based on the first training data set based on the second sub-training data set.
And taking the pre-constructed initial emotion classification model trained and finished based on the second sub-training data set as a text emotion classification model.
In this embodiment, in order to enable the text sentiment classification model to better process the target comment text to identify the sentiment tendency information contained in the target comment text, when performing model training, the training set for training the initial sentiment classification model includes a first sub-training data set and a second sub-training data set, and in order to improve the effect of training the initial sentiment classification model, the initial sentiment classification model is trained by using the first sub-training data set so that the initial sentiment classification model learns the comment sentence pattern contained in the sample comment data labeled with the first sentiment category, and after the initial sentiment classification model trained based on the first training data set, the initial sentiment classification model trained based on the first training data set is trained based on the second sub-training data set, therefore, the initial emotion classification model can learn the comment sentence pattern contained in the sample comment data labeled with the first emotion category more quickly based on the comment sentence pattern contained in the sample comment data labeled with the second emotion category, and the finally obtained text emotion classification model can better process the target comment text.
For example, in a specific application scenario, the sample comment data labeled with the first emotion category is the product comment data labeled with the first emotion category, and the sample comment data labeled with the second emotion category is the enterprise comment data for the enterprise and labeled with the emotion category. In order to enable the initial emotion classification model to better learn comment information for enterprises, the initial emotion classification model is trained by utilizing a first sub-training data set, and an initial comment information frame is learned. Then, training the initial emotion classification model trained and completed based on the first training data set based on the second sub-training data set, so that the initial emotion classification model can learn the comment information frame for the enterprise on the basis of the learned initial comment information frame, and after the initial emotion classification model is trained and completed based on the second sub-training data set, the target comment text for the enterprise can be better processed based on the text emotion classification model, and an emotion classification result corresponding to the target comment text is obtained.
S13: and determining the designated attribute parameters of the target object according to the emotion types of the target comment texts described in the emotion classification result.
As an example of the present application, the specified attribute parameter refers to data for presenting the condition of the target object. For example, when the attribute parameter is the competitiveness value of a business, the competitiveness of the business relative to other businesses in the market can be known through the competitiveness value.
The emotion category refers to emotional tendency information corresponding to the target comment text. For example, the emotion category of the target comment text pair is "positive".
It can be understood that, because the sentiment classification result includes the sentiment tendency information corresponding to the target comment text, and the sentiment tendency of the comment person on the target object can be known through the sentiment tendency information, the specified attribute parameter for describing the comment situation of the comment person on the target object can be determined according to the number of the target comment texts for the sentiment tendency of the target object described in the sentiment classification result.
With reference to fig. 2, in an embodiment of the present application, a plurality of target comment texts are provided, and the emotion classification result includes an emotion category corresponding to each of the plurality of target comment texts.
Determining the designated attribute parameters of the target object according to the emotion types of the target comment texts described in the emotion classification result, wherein the method comprises the following steps:
s21: and counting the number of the target comment texts respectively corresponding to each emotion type according to the emotion type respectively corresponding to each target comment text in the plurality of target comment texts.
S22: and determining the designated attribute parameters of the target object according to the number of the target comment texts respectively corresponding to each emotion category.
In the application, when a plurality of target comment texts are provided for the target object, and the emotional tendency information is recorded in each target comment text and corresponds to the emotional category, in order to better understand the overall emotional tendency of the target object for the reviewer, the number of the target comment texts corresponding to each emotional category can be counted, and the designated attribute parameters of the target object can be determined according to the number of the target comment texts corresponding to each emotional category.
Specifically, the designated attribute parameters of the target object are determined according to the ratio between the number of the target comment texts respectively corresponding to each emotion category and the total number of the target comment texts.
Illustratively, the emotion types comprise positive and negative, and the emotion types corresponding to the target comment text a, the target comment text B, the target comment text C and the target comment text D are respectively positive, negative and positive, and according to the number 3 of the target comment texts corresponding to the positive and the number 1 of the target comment texts corresponding to the negative, the number of the positive target comment texts is determined to be 75%, and further the specified attribute parameter of the target object is determined to be 75%.
In an embodiment of the present application, in order to better describe the target object, the specified attribute parameters of the target object are multiple, and the target object is described by the multiple specified attribute parameters.
In order to better determine the designated attribute parameters of the target object, before determining the designated attribute parameters of the target object according to the emotion classification of the target comment text described in the emotion classification result, the method further comprises the following steps:
and analyzing the plurality of target comment texts to obtain attribute categories corresponding to each target comment text.
As an example of the present application, an attribute category refers to a category corresponding to a certain reference attribute of a target object. For example, when the target object is an enterprise, the development innovation is to describe an attribute category of the enterprise.
It can be understood that, since the target comment text is a text formed according to comments made by reviewers for the target object, in order to better determine the specified attribute parameters of the target object based on the target comment text, a plurality of target comment texts are analyzed to obtain the attribute categories corresponding to each target comment text, so that when the emotion classification result of the target comment text is obtained subsequently, the attribute categories of the target comment text and the emotion classification of the target comment text can be corresponded to determine the specified attribute parameters of the target object.
The specific implementation mode for determining the designated attribute parameters of the target object according to the emotion types of the target comment texts described in the emotion classification result comprises the following steps:
and determining the designated attribute parameters of the target object according to the emotion classification results of the target comment texts associated with the corresponding attribute types and the designated attribute parameters.
In this embodiment, since the designated attribute parameters of the target object are multiple and each target comment text corresponds to one attribute type, in order to better determine each designated attribute parameter of the target object, the designated attribute parameters of the target object may be determined according to the emotion classification result of the target comment text associated with the designated attribute parameters and the corresponding attribute type.
It can be understood that, the specified attribute parameter of the target object is determined according to the emotion classification result of the target comment text associated with the specified attribute parameter by the corresponding attribute type, that is, when one specified attribute parameter of the target object is determined, each target comment text associated with the specified attribute parameter by the corresponding attribute type is determined first, so that the specified attribute parameter is determined according to the emotion classification result corresponding to each target comment text. Specifically, one designated attribute parameter of the target object is determined according to the ratio between the number of the target comment texts respectively corresponding to each emotion category and the total number of the target comment texts. And in the same way, all the specified attribute parameters of the target object are determined.
Illustratively, the target object includes 2 designated attribute parameters, which are designated attribute parameter a and designated attribute parameter B, respectively, and the attribute category associated with the designated attribute parameter a is C and the attribute category associated with the designated attribute parameter B is D. In addition, the target comment text 1, the target comment text 2, the target comment text 3, the target comment text 4 and the target comment text 5 are obtained, after each target comment text is analyzed, the attribute category corresponding to the target comment text 1, the target comment text 2 and the target comment text 3 is determined to be C, the attribute category corresponding to the target comment text 4 and the target comment text 5 is determined to be D, and meanwhile, the emotion classification results corresponding to the target comment text 1, the target comment text 2, the target comment text 3, the target comment text 4 and the target comment text 5 are positive, negative, positive and negative. And according to the emotion classification results corresponding to the target comment text 1, the target comment text 2 and the target comment text 3 which are associated with the corresponding attribute class C and the specified attribute parameter A, determining that the proportion of the positive emotion classification result is 66.66%, namely, the specified attribute parameter A is 66.66%, and so on, determining that the specified attribute parameter B is 50%.
In some embodiments, in order to better show the condition of the target object, after the designated attribute parameters of the target object are multiple and are determined according to the emotion classification result, a preset chart is generated according to each designated attribute parameter.
For example, the number of the designated attribute parameters of the target object is 5, which are designated attribute parameter a, designated attribute parameter B, designated attribute parameter C, designated attribute parameter D, and designated attribute parameter E, and the designated attribute parameter a is 10%, the designated attribute parameter B is 60%, the designated attribute parameter C is 40%, the designated attribute parameter D is 90%, and the designated attribute parameter E is 88%. Further, in order to visually represent the situation of the target object, a five-dimensional graph is generated according to the specified attribute parameter a, the specified attribute parameter B, the specified attribute parameter C, the specified attribute parameter D and the specified attribute parameter E, so that the specific aspect of the target object on the micro scale is represented, and the information on the macro scale is represented. Therefore, when the target object is an enterprise, the situation of the target object can be known clearly when one enterprise is analyzed and different enterprises are compared.
In an embodiment of the present application, in a possible scenario, a reviewer issues multiple pieces of repeated comment information for a target object, and if a specified attribute parameter of the target object is determined directly based on the repeated comment information, an error may exist, so a specific implementation manner of obtaining a target comment text for the target object includes:
a plurality of initial comment texts for a target object are acquired.
Deleting repeated comment texts in the plurality of initial comment texts according to the comment time, the information source, the communication address of the comment person and the identification information of the comment person, which are respectively corresponding to each initial comment text;
and determining a plurality of initial comment texts with the repeated comment texts deleted as target comment texts.
In the present embodiment, the comment time refers to a time point at which the comment information is posted by the reviewer for the target object. For example, 10 minutes at 8 months, 10 days and 10 days in 2021, the reviewer issues comment information about enterprise a, wherein the comment information is that research and development investment of the enterprise is large, research and development yield is high, and conversion application based on research and development innovation is small.
The information source refers to a platform for the reviewer to publish comment information for the target object. Such as forums, blogs, etc., in the area of the enterprise.
The communication address of the reviewer refers to an internet protocol address used by the reviewer when publishing comment information for the target object. For example, the reviewer issues review information to the enterprise a, and the reviewer has an Internet Protocol Address (IP Address) corresponding to the electronic device.
The identification information of the reviewer refers to the distinctive information corresponding to the target object to which the reviewer publishes the comment information, and the reviewer can be identified and distinguished through the distinctive information. Such as a nickname or account number of the reviewer.
In this embodiment, since there may be a case where the reviewer reviews the target object for multiple times, so that the specified attribute parameter is inaccurate when the specified attribute parameter of the target object is determined by using the information of the part of the repeated reviews, after obtaining a plurality of initial review texts for the target object, it is determined whether a repeated review text exists in the plurality of initial review texts through each information of the determined initial review text, and the repeated review text in the plurality of initial review texts is deleted, so that the plurality of initial review texts from which the repeated review text is deleted is determined as the target review text.
As a possible implementation manner of this embodiment, according to the comment time, the information source, the communication address of the reviewer, and the identification information of the reviewer, which correspond to each initial comment text, a specific implementation manner of deleting the repeated comment text in the multiple initial comment texts may include:
if more than two initial comment texts with the same communication address and different information sources exist in the initial comment texts, the hash values of the more than two initial comment texts are calculated respectively.
If the initial comment texts with the same hash value and the same identification information of the reviewers exist in more than two initial comment texts, the comment time difference value between every two initial comment texts with the same hash value and the same identification information of the reviewers is calculated.
If the minimum value of the comment time difference value between every two comment texts is smaller than the preset time length, one initial comment text is selected from the initial comment texts with the same hash value and the same identifier information of the reviewer for retention, and the rest of the initial comment texts are deleted.
In this embodiment, the preset time duration may be preset according to actual requirements.
It can be understood that, in the plurality of initial comment texts, more than two initial comment texts with the same communication address of the reviewer and different information sources exist, that is, the published comment information indicating that the reviewer may repeat for the target object is determined, so in order to determine whether the comment information is repeated, the hash value of each initial comment text in the initial comment texts with different information sources is determined, and whether the hash values of the initial comment texts are the same is determined. If two initial comment texts with the same hash value exist, that is, the two initial comment texts show that the same comment information is issued in different information sources for the target object, it is necessary to further determine whether the identification information of the reviewers corresponding to the initial comment texts with the same hash value is the same. When the identification information of the reviewers is the same, the comment information is repeatedly issued to the target object by the same reviewer. Then, whether the commentator maliciously publishes the comment information is determined, namely when two initial comment texts with the same identification information of the commentator exist is determined, a comment time difference value between comment times of every two initial comment texts is further determined, and when the comment time difference value is smaller than a preset time length, the fact that the commentator publishes repeated comment information for a target object in a short time is indicated, so that for the repeatedly published comment information, one initial comment text of at least two initial comment texts with the same identification information of the commentator needs to be used as the target comment text, and the rest of the initial comment texts are deleted.
It should be noted that, a manner of determining the hash value of each initial comment text published in the initial comment texts of different information sources may be referred to in the related art in the prior art, and details are not summarized here.
In some embodiments, a text passage for describing an object of a target is obtained, and a plurality of initial comment texts for the target object are obtained from the text passage.
The embodiment of the application can acquire and process related data based on an artificial intelligence technology. Among them, Artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.
The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.
Corresponding to the method for determining object attribute parameters in the foregoing embodiments, fig. 3 shows a block diagram of an apparatus for determining object attribute parameters provided in an embodiment of the present application, and for convenience of explanation, only the parts related to the embodiment of the present application are shown.
Referring to fig. 3, the apparatus 100 includes:
the obtaining module 101 is configured to, if it is detected that the terminal device is in a target operation state and an object selection instruction is received, obtain a target comment text for a target object according to an analysis result of the object selection instruction;
the processing module 102 is configured to perform emotion classification processing on the target comment text to obtain an emotion classification result of the target comment text;
and the determining module 103 is configured to determine the specified attribute parameter of the target object according to the emotion type of the target comment text described in the emotion classification result.
In an embodiment, the obtaining module 101 is further configured to obtain a text paragraph for describing the target object; target comment text is extracted from the text passage.
In an embodiment, the target comment texts are multiple, and the emotion classification result includes an emotion category corresponding to each target comment text in the multiple target comment texts.
The determining module 103 is further configured to count the number of target comment texts corresponding to each emotion category according to the emotion category corresponding to each target comment text in the plurality of target comment texts; and determining the designated attribute parameters of the target object according to the number of the target comment texts respectively corresponding to each emotion category.
In one embodiment, the apparatus 100 further comprises a parsing module.
The designated attribute parameters of the target object are multiple.
And the analysis module is used for analyzing the plurality of target comment texts to obtain attribute categories corresponding to each target comment text.
The determining module 103 is further configured to determine the specified attribute parameter of the target object according to the emotion classification result of the target comment text associated with the specified attribute parameter and the corresponding attribute type.
In one embodiment, the apparatus 100 further comprises a training module.
The processing module 102 is further configured to input the target comment text into the trained text emotion classification model for processing, so as to obtain an emotion classification result of the target comment text.
The training module is used for acquiring sample comment data marked with emotion types; and training to obtain a text emotion classification model by taking the sample comment data as a training set.
In an embodiment, the training set includes a first sub-training data set including sample opinion data labeled as a first sentiment category and a second sub-training data set including one or more of sample opinion data labeled as the first sentiment category or sample opinion data labeled as a second sentiment category.
And the training module is also used for optimizing and updating the parameters of the pre-constructed initial emotion classification model based on the first sub-training data set and the second sub-training data set, and determining the initial emotion classification model after parameter optimization and updating as the text emotion classification model.
The apparatus for determining object attribute parameters provided in this embodiment is used to implement any method for determining object attribute parameters in the method embodiments, where the functions of each module may refer to corresponding descriptions in the method embodiments, and the implementation principle and technical effect thereof are similar, and are not described herein again.
Fig. 4 is a schematic structural diagram of a terminal device according to an embodiment of the present application. As shown in fig. 4, the terminal device 4 of this embodiment includes: at least one processor 40 (only one processor is shown in fig. 4), a memory 41, and a computer program 42 stored in the memory 41 and executable on the at least one processor 40, the steps of any of the various method embodiments described above being implemented when the computer program 42 is executed by the processor 40.
The terminal device 4 may be a desktop computer, a notebook, a palm computer, a server, or other computing devices. The server can be an independent server, and can also be a cloud server for providing basic cloud computing services such as cloud service, a cloud database, cloud computing, network service, big data and an artificial intelligence platform.
The terminal device may include, but is not limited to, a processor 40, a memory 41. Those skilled in the art will appreciate that fig. 4 is merely an example of the terminal device 4, and does not constitute a limitation of the terminal device 4, and may include more or less components than those shown, or combine some components, or different components, such as an input-output device, a network access device, and the like.
The Processor 40 may be a Central Processing Unit (CPU), and the Processor 40 may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), off-the-shelf Programmable Gate arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 41 may in some embodiments be an internal storage unit of the terminal device 4, such as a hard disk or a memory of the terminal device 4. The memory 41 may be an external storage device of the terminal device 4 in other embodiments, such as a plug-in hard disk provided on the terminal device 4, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like. Further, the memory 41 may also include both an internal storage unit of the terminal device 4 and an external storage device. The memory 41 is used for storing an operating system, an application program, a BootLoader (BootLoader), data, and other programs, such as program codes of a computer program. The memory 41 may also be used to temporarily store data that has been output or is to be output.
It should be noted that, for the information interaction, execution process, and other contents between the above-mentioned devices/units, the specific functions and technical effects thereof are based on the same concept as those of the embodiment of the method of the present application, and specific reference may be made to the part of the embodiment of the method, which is not described herein again.
It will be clear to those skilled in the art that, for convenience and simplicity of description, the foregoing division of the functional modules is merely illustrated, and in practical applications, the above function distribution may be performed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to perform all or part of the above described functions. Each functional module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional modules are only used for distinguishing one functional module from another, and are not used for limiting the protection scope of the application. The specific working process of the modules in the apparatus may refer to the corresponding process in the foregoing method embodiment, and is not described herein again.
An embodiment of the present application further provides a terminal device, where the terminal device includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, the processor implementing the steps of any of the various method embodiments described above when executing the computer program.
The embodiments of the present application further provide a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments may be implemented.
The embodiments of the present application provide a computer program product, which when running on a terminal device, enables the terminal device to implement the steps in the above method embodiments when executed.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer readable storage medium and used by a processor to implement the steps of the embodiments of the methods described above. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to an apparatus/terminal device, recording medium, computer Memory, Read-Only Memory (ROM), Random-Access Memory (RAM), electrical carrier wave signals, telecommunications signals, and software distribution medium. Such as a usb-disk, a removable hard disk, a magnetic or optical disk, etc.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other ways. For example, the above-described embodiments of the apparatus/terminal device are merely illustrative, and for example, a module or a unit may be divided into only one logical function, and may be implemented in other ways, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.
Claims (10)
1. A method of determining object property parameters, comprising:
if the terminal equipment is detected to be in a target running state and an object selection instruction is received, acquiring a target comment text aiming at a target object according to an analysis result of the object selection instruction;
performing emotion classification processing on the target comment text to obtain an emotion classification result of the target comment text;
and determining the designated attribute parameters of the target object according to the emotion types of the target comment texts described in the emotion classification result.
2. The method of claim 1, wherein the target comment texts are multiple, and the emotion classification result contains an emotion category corresponding to each target comment text in the multiple target comment texts;
the determining the designated attribute parameters of the target object according to the emotion classification of the target comment text described in the emotion classification result includes:
counting the number of the target comment texts corresponding to each emotion type according to the emotion type corresponding to each target comment text in the plurality of target comment texts;
and determining the designated attribute parameters of the target object according to the number of the target comment texts respectively corresponding to each emotion category.
3. The method of claim 2, wherein the designated attribute parameters of the target object are plural;
before determining the designated attribute parameters of the target object according to the emotion classification of the target comment text described in the emotion classification result, the method further comprises the following steps:
analyzing the plurality of target comment texts to obtain attribute categories corresponding to each target comment text;
the determining the designated attribute parameters of the target object according to the emotion classification of the target comment text described in the emotion classification result includes:
and determining the designated attribute parameters of the target object according to the emotion classification result of the target comment text associated with the corresponding attribute type and the designated attribute parameters.
4. The method of claim 1, wherein the performing emotion classification processing on the target comment text to obtain an emotion classification result of the target comment text comprises:
inputting the target comment text into a trained text sentiment classification model for processing to obtain a sentiment classification result of the target comment text;
the text emotion classification model is obtained by training in the following mode:
obtaining sample comment data marked with emotion types;
and training to obtain the text emotion classification model by taking the sample comment data as a training set.
5. The method of claim 4, wherein the training set comprises a first sub-training data set comprising sample opinion data labeled as a first sentiment category and a second sub-training data set comprising one or more of sample opinion data labeled as a first sentiment category or sample opinion data labeled as a second sentiment category;
the training with the sample comment data as a training set to obtain the text emotion classification model comprises:
and optimizing and updating the parameters of the pre-constructed initial emotion classification model based on the first sub-training data set and the second sub-training data set, and determining the initial emotion classification model after parameter optimization and updating as the text emotion classification model.
6. The method of any one of claims 1-5, wherein the obtaining target comment text for a target object comprises:
acquiring a plurality of initial comment texts aiming at the target object;
deleting repeated comment texts in the plurality of initial comment texts according to comment time, information sources, communication addresses of the comments and identification information of the comments, which correspond to each initial comment text;
determining the plurality of initial comment texts with the repeated comment texts deleted as the target comment texts.
7. The method of claim 6, wherein deleting repeated comment texts in the plurality of initial comment texts according to the comment time, the information source, the communication address of the reviewer and the identification information of the reviewer corresponding to each initial comment text respectively comprises:
if more than two initial comment texts with the same communication address and different information sources exist in the plurality of initial comment texts, respectively calculating hash values of the more than two initial comment texts;
if the initial comment texts with the same hash values and the same identification information of the reviewers exist in the more than two initial comment texts, calculating the comment time difference value between every two initial comment texts with the same hash values and the same identification information of the reviewers;
and if the minimum value of the comment time difference value between every two comment texts is less than the preset time length, selecting one initial comment text from the initial comment texts with the same hash value and the same identifier information of the reviewer for retention, and deleting the rest initial comment texts.
8. An apparatus for determining object property parameters, comprising:
the acquisition module is used for acquiring a target comment text aiming at a target object according to an analysis result of an object selection instruction if the terminal equipment is detected to be in a target running state and the object selection instruction is received;
the processing module is used for executing emotion classification processing on the target comment text to obtain an emotion classification result of the target comment text;
and the determining module is used for determining the designated attribute parameters of the target object according to the emotion types of the target comment texts described in the emotion classification result.
9. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the method of any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which, when executed by a processor, implements the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111088965.7A CN113792145A (en) | 2021-09-16 | 2021-09-16 | Method and device for determining object attribute parameters, terminal equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111088965.7A CN113792145A (en) | 2021-09-16 | 2021-09-16 | Method and device for determining object attribute parameters, terminal equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113792145A true CN113792145A (en) | 2021-12-14 |
Family
ID=79183625
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111088965.7A Pending CN113792145A (en) | 2021-09-16 | 2021-09-16 | Method and device for determining object attribute parameters, terminal equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113792145A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105550269A (en) * | 2015-12-10 | 2016-05-04 | 复旦大学 | Product comment analyzing method and system with learning supervising function |
| US20180197192A1 (en) * | 2017-01-12 | 2018-07-12 | Hefei University Of Technology | Method and device for identifying preferential region of product |
| CN108446813A (en) * | 2017-12-19 | 2018-08-24 | 清华大学 | A kind of method of electric business service quality overall merit |
| CN113254647A (en) * | 2021-06-11 | 2021-08-13 | 大唐融合通信股份有限公司 | Course quality analysis method, device and system |
-
2021
- 2021-09-16 CN CN202111088965.7A patent/CN113792145A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105550269A (en) * | 2015-12-10 | 2016-05-04 | 复旦大学 | Product comment analyzing method and system with learning supervising function |
| US20180197192A1 (en) * | 2017-01-12 | 2018-07-12 | Hefei University Of Technology | Method and device for identifying preferential region of product |
| CN108446813A (en) * | 2017-12-19 | 2018-08-24 | 清华大学 | A kind of method of electric business service quality overall merit |
| CN113254647A (en) * | 2021-06-11 | 2021-08-13 | 大唐融合通信股份有限公司 | Course quality analysis method, device and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112860841B (en) | Text emotion analysis method, device, equipment and storage medium | |
| US20150142423A1 (en) | Phrase-based data classification system | |
| Harfoushi et al. | Sentiment analysis algorithms through azure machine learning: Analysis and comparison | |
| CN112070138B (en) | Construction method of multi-label mixed classification model, news classification method and system | |
| US9773252B1 (en) | Discovery of new business openings using web content analysis | |
| CN112163424A (en) | Data labeling method, device, equipment and medium | |
| CN109800354B (en) | Resume modification intention identification method and system based on block chain storage | |
| CN111324739B (en) | Text emotion analysis method and system | |
| CN110555451A (en) | information identification method and device | |
| CN114648392B (en) | Product recommendation method and device based on user portrait, electronic equipment and medium | |
| CN113704429A (en) | Semi-supervised learning-based intention identification method, device, equipment and medium | |
| CN111427974A (en) | Data quality evaluation management method and device | |
| CN111143505A (en) | Document processing method, device, medium and electronic equipment | |
| CN110069558A (en) | Data analysing method and terminal device based on deep learning | |
| CN113256181A (en) | Risk factor prediction method, device, equipment and medium | |
| CN116541528A (en) | Labeling method and system for recruitment field knowledge graph construction | |
| CN115470790A (en) | Method and device for identifying named entities in file | |
| CN113239273B (en) | Method, apparatus, device and storage medium for generating text | |
| CN112328812B (en) | Domain knowledge extraction method and system based on self-adjusting parameters and electronic equipment | |
| CN112115212B (en) | Parameter identification method and device and electronic equipment | |
| Singh et al. | Twitter sentiment analysis for stock prediction | |
| CN109242690A (en) | Finance product recommended method, device, computer equipment and readable storage medium storing program for executing | |
| CN114297380A (en) | Data processing method, device, equipment and storage medium | |
| CN113792145A (en) | Method and device for determining object attribute parameters, terminal equipment and storage medium | |
| CN113704599A (en) | Marketing conversion user prediction method and device and computer equipment |
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 |