CN109409529B - Event cognitive analysis method, system and storage medium - Google Patents
Event cognitive analysis method, system and storage medium Download PDFInfo
- Publication number
- CN109409529B CN109409529B CN201811069882.1A CN201811069882A CN109409529B CN 109409529 B CN109409529 B CN 109409529B CN 201811069882 A CN201811069882 A CN 201811069882A CN 109409529 B CN109409529 B CN 109409529B
- Authority
- CN
- China
- Prior art keywords
- data
- model
- label
- classification label
- internet
- 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.)
- Active
Links
Images
Classifications
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
- G06N3/08—Learning methods
Abstract
The invention relates to an event cognitive analysis method, a system and a storage medium, wherein the analysis method comprises the following steps: the method comprises the steps of obtaining internet data in real time, and processing the internet data to obtain a feature vector; establishing a multi-dimensional classification label library, and training based on the multi-dimensional classification label library and a machine learning algorithm to obtain a data label model; inputting the feature vectors into a data label model to obtain classification labels; and performing event cognition analysis based on the classification label. According to the embodiment of the invention, the characteristic vector of the internet data is obtained by processing the internet data, the data label model is obtained based on the machine learning algorithm, the characteristic vector is analyzed through the data label model to obtain the corresponding classification label, the internet data is intelligently processed to form the content relation network, and the event cognitive analysis is realized through the machine learning algorithm.
Description
Technical Field
The invention relates to the technical field of information analysis, in particular to an event cognitive analysis method, an event cognitive analysis system and a storage medium.
Background
With the rapid development of information technology and the rapid popularization of the internet, global data shows the characteristics of explosive growth and mass aggregation, and the expression form and the distribution channel of information are increasingly diversified.
Particularly, the media industry is generally building data centers, all operation modes taking news data as a center are close to industrial production of data information, the production of news is a brand new production mode with the adoption, writing, editing and storage in the front, and analysis and mining in the back, however, how to effectively organize and manage increasingly huge structured and unstructured data, fully integrate, mine and utilize rich information resources, expand new fields and new boundaries of information services, and become a core problem in the current information processing field.
Disclosure of Invention
In order to solve the problems in the prior art, at least one embodiment of the present invention provides an event awareness analysis method, system and storage medium.
In a first aspect, an embodiment of the present invention provides an event awareness analysis method, where the analysis method includes:
the method comprises the steps of obtaining internet data in real time, and processing the internet data to obtain a feature vector;
establishing a multi-dimensional classification label library, and training based on the multi-dimensional classification label library and a machine learning algorithm to obtain a data label model;
inputting the feature vector into a data label model to obtain a classification label; and performing event cognition analysis based on the classification label.
With reference to the first aspect, in a first embodiment of the first aspect, the processing the internet data to obtain a feature vector specifically includes:
carrying out structuralization processing on the internet data, and denoising to obtain structuralization data;
processing the structured data through a natural language processing technology to obtain standard data;
and selecting the characteristics of the standard data to obtain characteristic data, and vectorizing the characteristic data to obtain the characteristic vector.
With reference to the first embodiment of the first aspect, in a second embodiment of the first aspect, feature selection is performed on the standard data to obtain feature data, and the feature data is vectorized to obtain the feature vector;
acquiring the occurrence frequency of each prepared data in the standard data;
acquiring the occurrence times of corresponding prepared data in the pre-stored characteristic comparison data;
subtracting the occurrence frequency of corresponding preset data in the feature comparison data from the occurrence frequency of the prepared data in the standard data to obtain the feature frequency of the prepared data;
taking the prepared data with the characteristic times larger than a preset threshold value as the characteristic data;
and vectorizing all the feature data to obtain the feature vector.
With reference to the first aspect, in a third embodiment of the first aspect, the establishing a multidimensional classification label library, and obtaining a data label model based on the multidimensional classification label library and training of a machine learning algorithm specifically include:
establishing a multi-dimensional classification label library;
obtaining pre-stored model data, and adding a classification label to each model data based on the multi-dimensional classification label library;
processing each model data to obtain a model vector;
and training to obtain the data label model by taking the model vector of the model data as input and taking the classification label of the model data as output based on a machine learning algorithm.
With reference to the third embodiment of the first aspect, in a fourth embodiment of the first aspect, the multi-dimensional class label library comprises: at least two layers of classification labels;
each classification label layer includes at least one classification label.
With reference to the third embodiment of the first aspect, in a fifth embodiment of the first aspect, the machine learning algorithm comprises: vector machines, naive bayes algorithms, or convolutional neural networks.
With reference to the third embodiment of the first aspect, in a sixth embodiment of the first aspect, the training to obtain the data label model based on the machine learning algorithm with the model vector of the model data as an input and the classification label of the model data as an output specifically includes:
inputting the model vector of the model data into a convolutional neural network, and performing iteration according to a preset weight by utilizing the forward propagation of the convolutional neural network to obtain a prediction label;
calculating an error value of the prediction label and a classification label of the model data by utilizing back propagation of a convolutional neural network;
comparing the error value with a preset threshold value, and judging whether the error value is smaller than the preset threshold value;
when the error value is larger than or equal to a preset threshold value, adjusting the preset weight, and repeating the iteration to obtain a predicted label until the error value of the predicted label and the classification label is smaller than the preset threshold value;
or when the error value is smaller than a preset threshold value, obtaining the data label model.
With reference to the first aspect or any one of the first, second, third, fourth, fifth, or sixth embodiments of the first aspect, in a sixth embodiment of the first aspect, the obtaining internet data in real time specifically includes:
if the internet data is text data;
converting the Internet data of different language types into Internet data of a preset language type through language identification;
and processing the internet data to obtain the internet data without staying words and word segmentation.
In a second aspect, an embodiment of the present invention provides an event awareness analysis system, where the event awareness analysis system includes a processor and a memory; the processor is configured to execute the event awareness analysis program stored in the memory to implement the event awareness analysis method according to any one of the embodiments of the first aspect.
In a third aspect, an embodiment of the present invention provides a storage medium, where the storage medium stores one or more programs, and the one or more programs are executable by one or more processors to implement the event awareness analysis method described in any one of the first aspects.
Compared with the prior art, the technical scheme of the invention has the following advantages: according to the embodiment of the invention, the characteristic vector of the internet data is obtained by processing the internet data, the data label model is obtained based on the machine learning algorithm, the characteristic vector is analyzed through the data label model to obtain the corresponding classification label, the internet data is intelligently processed to form the content relation network, and the event cognitive analysis is realized through the machine learning algorithm.
Drawings
Fig. 1 is a schematic flow chart of an event cognition analysis method according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a method for event recognition analysis according to another embodiment of the present invention;
fig. 3 is a first schematic flow chart of an event awareness analysis method according to another embodiment of the present invention;
FIG. 4 is a second flowchart of an event awareness analysis method according to another embodiment of the present invention;
FIG. 5 is a third schematic flow chart of an event recognition analysis method according to another embodiment of the present invention;
fig. 6 is a schematic structural diagram of an event awareness analysis system according to yet another embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
As shown in fig. 1, an event cognition analysis method provided in an embodiment of the present invention includes:
and S11, acquiring the internet data in real time, and processing the internet data to obtain the feature vector.
In this embodiment, the internet data includes: text, pictures and various data of different types are not described herein again. Different types of data such as characters and pictures can be converted into corresponding feature vectors to express the data, so that the data is convenient for a user to process. If the internet data is a picture, the picture is converted into a feature vector, a gray image of the picture can be obtained, feature pixel points are obtained according to the gray image, and then the feature vector is constructed by the feature pixel points. If the internet data is text data, processing the text data in the internet to obtain a feature vector corresponding to the text data, and constructing the feature vector based on the extracted feature data by performing feature extraction on the text data. The feature extraction algorithm comprises the following steps: the TF-IDF algorithm and the chi-square algorithm. TF-IDF (term frequency-inverse document frequency) is a statistical method, a commonly used weighting technique for information retrieval and data mining, and words that can be used as feature data are obtained by performing data mining on text data to evaluate the importance degree of a word to one of documents in a document set or a corpus. The importance of a word increases in proportion to the number of times it appears in a document, but at the same time decreases in inverse proportion to the frequency with which it appears in the corpus. The main idea of TF-IDF is: if a word or phrase appears in an article with a high frequency TF and rarely appears in other articles, the word or phrase is considered to have a good classification capability and is suitable for classification. The chi-square algorithm quantifies the weight of the feature words from the category angle, and the best classification effect can be obtained by using the chi-square as the algorithm in the feature selection process from the actual experiment.
In this embodiment, because the format and the size of each type of data of the internet data are not consistent, after the internet data is obtained, the internet data can be processed, so that the formats of the different types of data are consistent, and the processing is convenient, as shown in fig. 2, the method for processing the internet data to obtain the feature vector includes:
and S21, carrying out structuralization processing on the internet data, and denoising to obtain structuralization data.
In this embodiment, the internet data is structured to accelerate the subsequent processing of the internet data, thereby improving the working efficiency and reducing the interference items, wherein the structured processing means: and adjusting the same type of internet data according to a preset structure rule to enable the same type of internet data to have a uniform data structure. For example, if the internet data is text data, the title, the release time, the author, and the body of the text are sequentially distributed, so that the data processing process can be rapidly and accurately performed, if the internet data is a picture, the resolution of the internet data is adjusted to be consistent, the pattern recognition is convenient, and other types of internet data can be adjusted according to any attribute of the internet data, so that the structure of the internet data has certain consistency, and the processing is convenient.
In this embodiment, in the text data, the text data in different languages is likely to cause data loss or analysis errors in the data processing process, and the time consumption of the text data in different languages in the data processing process is longer, so that the data processing efficiency is not high.
If the internet data is text data; converting the Internet data of different language types into Internet data of a preset language type through language identification; the language identification technology, such as various translation software, can be realized, and text data of different language types are converted into text data of the same language type.
The internet data is processed to obtain the internet data for removing the staying words and performing word segmentation, and most of the staying words in the text data play a role in connecting the context, so that the staying words in the text data can be removed. The text data can be segmented according to words by word segmentation, and the preprocessed internet data is obtained.
And S22, processing the structured data through a natural language processing technology to obtain standard data.
In the embodiment, the natural language processing technology relates to the fields of computer science and artificial intelligence; the collected data is preliminarily processed by a natural language processing technology, and standardization and knowledge are carried out.
And S23, performing feature selection on the standard data to obtain feature data, and vectorizing the feature data to obtain a feature vector.
In this embodiment, if the internet data is a picture, the picture is converted into the feature vector, the gray scale image of the picture is obtained, the feature pixel points are obtained according to the gray scale image, and then the feature vector is constructed by the feature pixel points. If the internet data is text data, processing the text data in the internet to obtain a feature vector corresponding to the text data, and specifically, constructing the feature vector based on the extracted feature data by performing feature extraction on the text data.
And S12, establishing a multi-dimensional classification label library, and training based on the multi-dimensional classification label library and a machine learning algorithm to obtain a data label model.
In this embodiment, Machine Learning (ML) is a multi-domain cross subject, and relates to multiple subjects such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, and the like; the special research on how a computer simulates or realizes the learning behavior of human beings so as to acquire new knowledge or skills and reorganize the existing knowledge structure to continuously improve the performance of the computer. In the step, the internal relation between the internet data and the classification label is summarized, and a corresponding data label model is generated.
Specifically, as shown in fig. 3, the method for obtaining the data label model based on the machine learning algorithm training includes:
and S31, establishing a multi-dimensional classification label library.
In this embodiment, the multidimensional classification tag library includes five-level, nearly eight thousand high-dimensional tags, and combines professional news topics and domain subject knowledge; the multi-dimensional classification label library may be manually input by the user or may be obtained by the user based on past experience. The multi-dimensional classification label library comprises: at least two layers of classification labels; each classification label layer includes at least one classification label. The classification label in the previous classification label layer is an upper classification label of the next classification label layer, for example, the first classification label layer comprises food, the next classification label layer can comprise staple food, meat, vegetables and the like, and the next classification label layer can be specific to which kind of meat or which kind of vegetables, the data analysis efficiency is improved by setting the hierarchical classification label. And extracting the common content by adopting a multiple division method, and reducing the space of the category list by adopting a known subdivision structure of an imitation division means.
And S32, obtaining pre-stored model data, and adding a classification label to each model data based on the multi-dimensional classification label library.
In this embodiment, pre-stored model data is obtained, classification models are respectively added to the model data, classification labels can be added through data analysis, and classification labels can also be manually added to each model data by people.
And S33, processing each model data to obtain a model vector.
In the present embodiment, model vectors are obtained from model data by processing the model data in the same manner as the above-described internet data.
And S34, training to obtain a data label model by taking the model vector of the model data as input and the classification label of the model data as output based on the machine learning algorithm.
In the embodiment, the internal relation between the internet data and the classification labels is summarized, the corresponding data label model is generated, a model is provided for automatic analysis of the internet data subsequently, and the rapid analysis is realized for adding the labels to the internet data.
Specifically, as shown in fig. 4, the implementation method of step S34 includes:
and S41, inputting the model vector of the model data into the convolutional neural network, and performing iteration according to the preset weight by utilizing the forward propagation of the convolutional neural network to obtain the prediction label.
And taking the model vector as the input of a convolutional neural network, iterating the convolutional neural network through chain conduction, and convolving a plurality of complete matrixes to obtain an output and a corresponding prediction label.
And S42, calculating an error value of the prediction label and the classification label of the model data by utilizing the back propagation of the convolutional neural network.
And performing back propagation on the convolutional neural network according to the classification label as output to calculate the error value of the prediction label and the classification label.
And S43, comparing the error value with a preset threshold value, judging whether the error value is smaller than the preset threshold value, and judging whether the processing of the convolutional neural network is stable and accurate by judging whether the error value is smaller than the preset threshold value.
And S44a, when the error value is greater than or equal to the preset threshold value, adjusting the preset weight, and repeating the iteration to obtain the predicted label until the error value between the predicted label and the classification label is less than the preset threshold value.
In this embodiment, the preset weight of each propagation node in the convolutional neural network is adjusted according to the error value, and the above process is continuously repeated to continuously adjust the convolutional neural network, thereby improving the accuracy of data identification.
S44b, or when the error value is smaller than the preset threshold value, obtaining the data label model.
In this embodiment, the model vector is used as an input of the convolutional neural network, the convolutional neural network iterates through chain conduction, and convolves a plurality of complete matrices to obtain an output and a corresponding prediction label, the convolutional neural network is subjected to back propagation to calculate an error value between the prediction label and the classification label according to the classification label as the output, a preset weight of each propagation node in the convolutional neural network is adjusted according to the error value, and the process of training the convolutional neural network is implemented by subjecting the plurality of model vectors and the corresponding classification labels to the above process.
S13, inputting the feature vectors into the data label model to obtain classification labels; and performing event cognition analysis based on the classification label.
In this embodiment, the feature vectors are analyzed based on the data label model to obtain classification labels, the classification labels are added to the internet data, and the labels have different levels, so that the internet data can be used for performing entity identification, relationship extraction, emotion analysis, contrastive analysis, time extraction, viewpoint analysis and other deep analysis after having the classification labels, thereby realizing multi-dimensional cognitive analysis of time, and enabling the analysis to be more flexible, accurate and efficient.
In this embodiment, the machine learning algorithm includes: vector machines, naive bayes algorithms, or convolutional neural networks.
As shown in fig. 5, an event cognition analysis method according to an embodiment of the present invention is different from the analysis method shown in fig. 1 in that:
and S51, acquiring the occurrence number of each prepared data in the standard data.
In this embodiment, the occurrence frequency of each prepared data in the standard data after being processed by the internet data, such as any word in the text data or a pixel point of any gray value in the picture, is obtained.
And S52, acquiring the occurrence times of corresponding prepared data in the pre-stored characteristic comparison data.
The pre-stored feature comparison data is text data or pictures of any type, and in this embodiment, data of corresponding words or the number of corresponding pixel points in the feature comparison data is obtained.
And S53, subtracting the occurrence frequency of the corresponding preset data in the feature comparison data from the occurrence frequency of the prepared data in the standard data to obtain the feature frequency of the prepared data.
And subtracting the times of the words or the pixel points appearing in other texts or pictures from the number of the words or the pixel points in the standard data to obtain the characteristic times of the prepared data.
And S54, taking the prepared data with the characteristic times larger than a preset threshold value as characteristic data.
In this embodiment, the feature times are compared with a preset threshold, and if the feature times are larger, it can be known that the occurrence times of the preliminary data in the standard data are higher and the occurrence times of the preliminary data in the other text data are smaller, so that the preliminary data can be used as the feature data.
And S55, vectorizing all the feature data to obtain feature vectors.
As shown in fig. 6, an embodiment of the present invention provides an event awareness analysis system, which includes a processor and a memory; the processor is used for executing the event cognition analysis program stored in the memory so as to realize the event cognition analysis method of any one of the embodiments.
The storage medium for recording the program code of the software program that can realize the functions of the above-described embodiments is provided to the system or apparatus in the above-described embodiments, and the program code stored in the storage medium is read and executed by the computer (or CPU or MPU) of the system or apparatus.
In this case, the program code itself read out from the storage medium performs the functions of the above-described embodiments, and the storage medium storing the program code constitutes an embodiment of the present invention.
As a storage medium for supplying the program code, for example, a flexible disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD-R, magnetic tape, nonvolatile memory card, ROM, and the like can be used.
The functions of the above-described embodiments may be realized not only by executing the readout program code by the computer, but also by some or all of actual processing operations executed by an OS (operating system) running on the computer according to instructions of the program code.
Further, the embodiments of the present invention also include a case where after the program code read out from the storage medium is written into a function expansion card inserted into the computer or into a memory provided in a function expansion unit connected to the computer, a CPU or the like included in the function expansion card or the function expansion unit performs a part of or the whole of the processing in accordance with the command of the program code, thereby realizing the functions of the above-described embodiments.
The embodiment of the present invention provides a storage medium, where one or more programs are stored in the storage medium, and the one or more programs may be executed by one or more processors to implement the event awareness analysis method according to any one of the embodiments of the first aspect.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will 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; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (8)
1. An event awareness analysis method, the analysis method comprising:
the method comprises the steps of obtaining internet data in real time, and processing the internet data to obtain a feature vector; the Internet data is text data;
establishing a multi-dimensional classification label library, and training based on the multi-dimensional classification label library and a machine learning algorithm to obtain a data label model;
inputting the feature vector into a data label model to obtain a classification label; performing event-aware analysis based on the classification label;
the establishing of the multi-dimensional classification label library and the training of the machine learning algorithm based on the multi-dimensional classification label library to obtain the data label model specifically comprise the following steps:
establishing a multi-dimensional classification label library;
obtaining pre-stored model data, and adding a classification label to each model data based on the multi-dimensional classification label library;
processing each model data to obtain a model vector;
based on a machine learning algorithm, taking a model vector of the model data as input, taking a classification label of the model data as output, and training to obtain the data label model;
the machine learning algorithm-based method includes the following steps of taking a model vector of model data as input and taking a classification label of the model data as output, and training to obtain the data label model, and specifically includes:
inputting the model vector of the model data into a convolutional neural network, and performing iteration according to a preset weight by utilizing the forward propagation of the convolutional neural network to obtain a prediction label;
calculating an error value of the prediction label and a classification label of the model data by utilizing back propagation of a convolutional neural network;
comparing the error value with a preset threshold value, and judging whether the error value is smaller than the preset threshold value;
when the error value is larger than or equal to a preset threshold value, adjusting the preset weight, and repeating the iteration to obtain a predicted label until the error value of the predicted label and the classification label is smaller than the preset threshold value;
or when the error value is smaller than a preset threshold value, obtaining the data label model.
2. The event awareness analysis method according to claim 1, wherein the processing the internet data to obtain a feature vector specifically includes:
carrying out structuralization processing on the internet data, and denoising to obtain structuralization data;
processing the structured data through a natural language processing technology to obtain standard data;
and selecting the characteristics of the standard data to obtain characteristic data, and vectorizing the characteristic data to obtain the characteristic vector.
3. The event awareness analysis method according to claim 2, wherein feature selection is performed on the standard data to obtain feature data, and the feature data is vectorized to obtain the feature vector;
acquiring the occurrence frequency of each prepared data in the standard data;
acquiring the occurrence times of corresponding prepared data in the pre-stored characteristic comparison data;
subtracting the occurrence frequency of corresponding preset data in the feature comparison data from the occurrence frequency of the prepared data in the standard data to obtain the feature frequency of the prepared data;
taking the prepared data with the characteristic times larger than a preset threshold value as the characteristic data;
and vectorizing all the feature data to obtain the feature vector.
4. The event awareness analysis method of claim 1, wherein the multi-dimensional class label library comprises: at least two layers of classification labels;
each classification label layer includes at least one classification label.
5. The event awareness analysis method of claim 1, wherein the machine learning algorithm comprises: vector machines, naive bayes algorithms, or convolutional neural networks.
6. The event awareness analysis method according to any one of claims 1 to 5, wherein the obtaining internet data in real time specifically comprises:
if the internet data is text data;
converting the Internet data of different language types into Internet data of a preset language type through language identification;
and processing the internet data to obtain the internet data without staying words and word segmentation.
7. An event awareness analysis system, comprising a processor, a memory; the processor is used for executing the event cognition analysis program stored in the memory so as to realize the event cognition analysis method of any claim 1-6.
8. A storage medium storing one or more programs executable by one or more processors to implement the event awareness analysis method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811069882.1A CN109409529B (en) | 2018-09-13 | 2018-09-13 | Event cognitive analysis method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811069882.1A CN109409529B (en) | 2018-09-13 | 2018-09-13 | Event cognitive analysis method, system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109409529A CN109409529A (en) | 2019-03-01 |
| CN109409529B true CN109409529B (en) | 2020-12-08 |
Family
ID=65464872
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811069882.1A Active CN109409529B (en) | 2018-09-13 | 2018-09-13 | Event cognitive analysis method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109409529B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111666270A (en) * | 2020-06-03 | 2020-09-15 | 北京软通智慧城市科技有限公司 | Event analysis system and event analysis method |
| CN113139141B (en) * | 2021-04-22 | 2023-10-31 | 康键信息技术（深圳）有限公司 | User tag expansion labeling method, device, equipment and storage medium |
| CN113779343A (en) * | 2021-09-18 | 2021-12-10 | 北京锐安科技有限公司 | Mass data processing method, device, medium and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103703487A (en) * | 2011-07-25 | 2014-04-02 | 国际商业机器公司 | Information identification method, program and system |
| CN107341687A (en) * | 2017-06-01 | 2017-11-10 | 华南理工大学 | A kind of proposed algorithm based on more dimension labels and classification and ordination |
| CN107622333A (en) * | 2017-11-02 | 2018-01-23 | 北京百分点信息科技有限公司 | A kind of event prediction method, apparatus and system |
| US10019983B2 (en) * | 2012-08-30 | 2018-07-10 | Aravind Ganapathiraju | Method and system for predicting speech recognition performance using accuracy scores |
| CN108304468A (en) * | 2017-12-27 | 2018-07-20 | 中国银联股份有限公司 | A kind of file classification method and document sorting apparatus |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101719121A (en) * | 2009-11-19 | 2010-06-02 | 上海第二工业大学 | Method of term weighing combination based on principal component analysis |
| US10438121B2 (en) * | 2014-04-30 | 2019-10-08 | International Business Machines Corporation | Automatic construction of arguments |
| CN106250513B (en) * | 2016-08-02 | 2021-04-23 | 西南石油大学 | Event modeling-based event personalized classification method and system |
| CN106447383A (en) * | 2016-08-30 | 2017-02-22 | 杭州启冠网络技术有限公司 | Cross-time multi-dimensional abnormal data monitoring method and system |
-
2018
- 2018-09-13 CN CN201811069882.1A patent/CN109409529B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103703487A (en) * | 2011-07-25 | 2014-04-02 | 国际商业机器公司 | Information identification method, program and system |
| US10019983B2 (en) * | 2012-08-30 | 2018-07-10 | Aravind Ganapathiraju | Method and system for predicting speech recognition performance using accuracy scores |
| CN107341687A (en) * | 2017-06-01 | 2017-11-10 | 华南理工大学 | A kind of proposed algorithm based on more dimension labels and classification and ordination |
| CN107622333A (en) * | 2017-11-02 | 2018-01-23 | 北京百分点信息科技有限公司 | A kind of event prediction method, apparatus and system |
| CN108304468A (en) * | 2017-12-27 | 2018-07-20 | 中国银联股份有限公司 | A kind of file classification method and document sorting apparatus |
Non-Patent Citations (3)
| Title |
|---|
| Spectral Classification Using Restricted Boltzmann Machine;Chen Fuqiang,Wu Yan,Bu Yude,Zhao Guodong;《Publications of the Astronomical Society of Australia 》;20141231;第198-203页 * |
| 基于多维扩展特征与深度学习的微博短文本情感分析;孙晓;《电子与信息学报》;20170930;第2048页--2055页 * |
| 深度学习应用于网络空间安全的现状、趋势与展望;张玉清;《计算机研究与发展》;20180130;第1117-1138页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109409529A (en) | 2019-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110309331B (en) | Cross-modal deep hash retrieval method based on self-supervision | |
| CN107705066B (en) | Information input method and electronic equipment during commodity warehousing | |
| CN109271521B (en) | Text classification method and device | |
| US9336299B2 (en) | Acquisition of semantic class lexicons for query tagging | |
| CN110196980B (en) | Domain migration on Chinese word segmentation task based on convolutional network | |
| CN114416927A (en) | Intelligent question and answer method, device, equipment and storage medium | |
| CN110852110B (en) | Target sentence extraction method, question generation method, and information processing apparatus | |
| CN109409529B (en) | Event cognitive analysis method, system and storage medium | |
| CN110083729B (en) | Image searching method and system | |
| KR20200127020A (en) | Computer-readable storage medium storing method, apparatus and instructions for matching semantic text data with tags | |
| CN113961685A (en) | Information extraction method and device | |
| CN112101031B (en) | Entity identification method, terminal equipment and storage medium | |
| CN110968725B (en) | Image content description information generation method, electronic device and storage medium | |
| CN113392179A (en) | Text labeling method and device, electronic equipment and storage medium | |
| CN114510570A (en) | Intention classification method and device based on small sample corpus and computer equipment | |
| CN113010678A (en) | Training method of classification model, text classification method and device | |
| CN115270797A (en) | Text entity extraction method and system based on self-training semi-supervised learning | |
| CN117635275A (en) | Intelligent electronic commerce operation commodity management platform and method based on big data | |
| CN110309355A (en) | Generation method, device, equipment and the storage medium of content tab | |
| CN116049376B (en) | Method, device and system for retrieving and replying information and creating knowledge | |
| CN114970553B (en) | Information analysis method and device based on large-scale unmarked corpus and electronic equipment | |
| CN112487263A (en) | Information processing method, system, equipment and computer readable storage medium | |
| Devgan et al. | Semi‐Supervised Learning | |
| CN115600595A (en) | Entity relationship extraction method, system, equipment and readable storage medium | |
| CN114842301A (en) | Semi-supervised training method of image annotation 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |