CN113326372A - Intellectual property data analysis method based on technical position - Google Patents
Intellectual property data analysis method based on technical position Download PDFInfo
- Publication number
- CN113326372A CN113326372A CN202110525628.3A CN202110525628A CN113326372A CN 113326372 A CN113326372 A CN 113326372A CN 202110525628 A CN202110525628 A CN 202110525628A CN 113326372 A CN113326372 A CN 113326372A
- Authority
- CN
- China
- Prior art keywords
- technical
- intellectual property
- property data
- similarity
- enterprise
- 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 30
- 238000007405 data analysis Methods 0.000 title claims abstract description 15
- 239000011159 matrix material Substances 0.000 claims abstract description 14
- 238000004458 analytical method Methods 0.000 claims abstract description 11
- 238000005516 engineering process Methods 0.000 claims description 27
- 230000006870 function Effects 0.000 claims description 15
- 238000003058 natural language processing Methods 0.000 claims description 8
- 230000000694 effects Effects 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000007477 logistic regression Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000011161 development Methods 0.000 abstract description 8
- 230000018109 developmental process Effects 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000035784 germination Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training 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
- G06F16/358—Browsing; Visualisation therefor
-
- 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
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- 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
- G06F16/355—Class or cluster creation or modification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to the technical field of intellectual property data analysis, in particular to an intellectual property data analysis method based on technical positions; the method comprises the following steps: step 1: the method comprises the steps of butting with a patent database of the national intellectual property office through an API (application programming interface), inputting a name of a user enterprise to obtain a patent of the user enterprise and storing the patent in a cloud database; step 2: indexing the technical and efficacy indexes of the specifications in the patents of the user enterprises to obtain a technical feature and achieved technical and efficacy set, and constructing a technical function matrix; and step 3: calculating the technical similarity between the patent and the specification by using a cosine included angle method; and 4, step 4: performing patent clustering analysis to obtain a clustering result; and 5: constructing a life cycle curve model; step 6: and combining the clustering result with the life cycle curve model to obtain an intellectual property data analysis result. The invention can intuitively know the technical development and the patent layout condition of the enterprise and prospectively know the future technical development and the patent layout direction of the enterprise.
Description
Technical Field
The invention relates to the technical field of intellectual property data analysis, in particular to an intellectual property data analysis method based on technical positions.
Background
At present, masses are more and more strong to intellectual property's protection consciousness, and to the science and technology enterprise, the patent is the effective means of protection independent innovation, however, among the prior art, the enterprise mainly through collect the patent after applying for, the manual work carries out analysis processes, and the power is spent consuming time, lacks the effective means that carries out intelligent analysis to the patent for the enterprise lacks the guidance on carrying out patent layout and technical development direction.
Therefore, the intellectual property data analysis method based on the technical position is provided.
Disclosure of Invention
The invention aims to provide an intellectual property data analysis method based on technical positions, which intuitively displays the correlation and connection of all patent technologies and technical hotspots and blank spots in a specific field in a period from establishment of an enterprise to current time of the enterprise through a matrix diagram, so that the enterprise integrally knows the technical development and the patent layout situation of the enterprise, prospectively knows the future technical development and the patent layout direction of the enterprise, and improves the reliability and the practicability of the patent layout.
In order to achieve the purpose, the technical scheme of the invention is as follows:
an intellectual property data analysis method based on technical positions comprises the following steps:
step 1: the method comprises the steps of butting with a patent database of the national intellectual property office through an API (application programming interface), inputting a name of a user enterprise to obtain a patent of the user enterprise and storing the patent in a cloud database;
step 2: indexing the technical and efficacy indexes of the specifications in the patents of the user enterprises to obtain a technical feature and achieved technical and efficacy set, and constructing a technical function matrix;
and step 3: calculating the technical similarity between the patent and the specification by using a cosine included angle method;
and 4, step 4: performing patent clustering analysis to obtain a clustering result;
and 5: constructing a life cycle curve model;
step 6: and combining the clustering result with the life cycle curve model to obtain an intellectual property data analysis result.
Specifically, the step 2 of indexing the claims in the user enterprise patent includes the following specific steps: performing NLP natural language processing on the specifications in the patents of the user enterprises, performing word segmentation, word stop removal and word frequency statistics, extracting technical keywords and achieved technical effects, ranking according to word frequency, and converting into a technical function matrix Csj＝{(T11，E11)，(T12，E12)，...，(Tsj，Esj)，...，(Twn，Ewn) A, (j ═ 1, 2, 3.., n), (s ═ 1, 2, 3.., w), where T is presentsjShowing the technical characteristics in patent s, EsjShowing the efficacy of the corresponding technical features in patent s.
Specifically, the formula of the cosine included angle in the step 3 is
Wherein Ai is composed of { kCa1，kCa2，kCa3，...，kCaj，...，kCanComposed of { kC }, Bib1，kCb2，kCb3，...，kCbj，...，kCbnIn which, kCajIs (T)aj，Eaj) The number of occurrences in patent a, (a ═ 1, 2, 3,. eta., w), kCbjIs (T)bj，Ebj) The number of occurrences in patent b, (b ═ 1, 2, 3.,w), when the similarity is 0, namely the similarity of the patent a and the patent b is the maximum, and when the similarity is gradually increased, namely the similarity of the patent a and the patent b is gradually decreased, the previous steps are repeated and the similarity value is converted into a similarity matrix.
Specifically, the proprietary clustering analysis method in the step 4 is a k-medoids algorithm, and the steps are as follows:
(41) classifying the patents of the user enterprises according to the IPC classification number to obtain q total classes;
(42) randomly selecting r classes from q total classes as reference points Zt(t＝1，2，3，...，r)；
(43) Distributing the remaining q-r technical subjects to each cluster class according to the principle closest to medoids;
(44) for all classes except the corresponding reference point in the t-th cluster class, sequentially calculating the value of the criterion function when the class is a new reference point, traversing all possibilities, and selecting the corresponding class when the criterion function is minimum as a new cluster class;
(45) repeating the processes (43) and (44) until all classes are not changed or the set maximum iteration number is reached;
(46) finally, r cluster classes are determined.
Specifically, the step of constructing the life cycle curve model in the step 5 is as follows: from establishment of an enterprise to the present, model construction is carried out on the related application amount and the related patent application amount of a certain technology, wherein time is an X axis, the number of patent applications is a Y axis, a two-dimensional S curve is drawn, logistic regression fitting is carried out on the two-dimensional S curve, and the life cycle of the technology is obtained in a quantitative and qualitative combination mode according to preset technology growth rate and technology maturity coefficient evaluation indexes.
The invention has the beneficial effects that: the invention takes technology, function and life cycle as research objects, constructs a patent technology function matrix through a technology function matrix analysis method, can visually present analysis results, and reflects technology correlation and relation in a matrix diagram, thereby concisely and briefly presenting technical characteristics and conditions, more comprehensively displaying technical characteristic contents, and simultaneously finding blank points, hot points and periodic points of technical efficacy, and developing construction and layout thinking of a patent network according to actual requirements and technical conditions; the clustering result is combined with the life cycle curve model, so that the reliability and the practicability of patent layout can be improved.
Drawings
FIG. 1 is a flow chart of a method according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a method for analyzing intellectual property data based on technical position includes the following steps:
step 1: the method comprises the steps of butting with a patent database of the national intellectual property office through an API (application programming interface), inputting a name of a user enterprise to obtain a patent of the user enterprise and storing the patent in a cloud database;
step 2: indexing the technical and efficacy indexes of the specifications in the patents of the user enterprises to obtain a technical feature and achieved technical and efficacy set, and constructing a technical function matrix;
and step 3: calculating the technical similarity between the patent and the specification by using a cosine included angle method;
and 4, step 4: performing patent clustering analysis to obtain a clustering result;
and 5: constructing a life cycle curve model;
step 6: and combining the clustering result with the life cycle curve model to obtain an intellectual property data analysis result.
Furthermore, the cloud database related by the invention configures a corresponding sub-database for each user enterprise, the preset patent classification in the sub-database comprises invention patents and utility model patents, and the corresponding types of patent data are separated from the cloud database and stored in the corresponding sub-databases.
Specifically, the step 2 of indexing the claims in the user enterprise patent includes the following specific steps: performing NLP natural language processing on the specifications in the patents of the user enterprises, performing word segmentation, word stop removal and word frequency statistics, extracting technical keywords and achieved technical effects, ranking according to word frequency, and converting into a technical function matrix Csj＝{(T11，E11)，(T12，E12)，...，(Tsj，Esj)，...，(Twn，Ewn) A, (j ═ 1, 2, 3.., n), (s ═ 1, 2, 3.., w), where T is presentsjShowing the technical characteristics in patent s, EsjShowing the efficacy of the corresponding technical features in patent s.
Furthermore, the NLP natural language processing method related by the invention is that a patent specification is read according to a patent science and technology dictionary, a general dictionary and an industry dictionary; the method comprises the steps of carrying out vector training by adopting a word vector technology, removing useless characters, stop words and low-frequency words in the patent specification, labeling the patent specification according to preset semantic rules, extracting technical keywords and achieving technical effects, wherein the preset semantic rules are W (R, L1 and L2), L1 and L2 respectively represent prefixes and suffixes of semantics, and R represents occurrence rules of the semantics.
Furthermore, the technical efficacy semantics of the invention have the remarkable characteristics that: (1) the beneficial effect positions of the specification are often concentrated and usually show that a plurality of clauses are connected together; (2) and (5) fixing and matching. Such as: adopted. (3) The same application authors have written their writings with certain regularity and similarity.
Specifically, the formula of the cosine included angle in the step 3 is
Wherein Ai is composed of { kCa1，kCa2，kCa3，...，kCaj，...，kCanComposed of { kC }, Bib1，kCb2，kCb3，...，kCbj，...，kCbnIn which, kCajIs (T)aj，Eaj) The number of occurrences in patent a, (a ═ 1, 2, 3,. eta., w), kCbjIs (T)bj，Ebj) In the number of occurrences in patent b, (b ═ 1, 2, 3., w), when similarity is 0, i.e., the degree of similarity between patent a and patent b is maximum, and when similarity becomes gradually larger, i.e., the degree of similarity between patent a and patent b becomes gradually smaller, the previous steps are repeated and the similarity values are converted into a similarity matrix.
Furthermore, the invention needs to normalize Ai and Bi when calculating the cosine included angle, and after data are normalized, the optimization process of the optimal solution becomes gentle, and the optimal solution can be converged more easily and correctly.
Specifically, the proprietary clustering analysis method in the step 4 is a k-medoids algorithm, and the steps are as follows:
(47) classifying the patents of the user enterprises according to the IPC classification number to obtain q total classes;
(48) randomly selecting r classes from q total classes as reference points Zt(t＝1，2，3，...，r)；
(49) Distributing the remaining q-r technical subjects to each cluster class according to the principle closest to medoids;
(50) for all classes except the corresponding reference point in the t-th cluster class, sequentially calculating the value of the criterion function when the class is a new reference point, traversing all possibilities, and selecting the corresponding class when the criterion function is minimum as a new cluster class;
(51) repeating the processes (43) and (44) until all classes are not changed or the set maximum iteration number is reached;
(52) finally, r cluster classes are determined.
Furthermore, compared with the k-means algorithm, the k-means algorithm adopted by the invention has smaller sensitivity to abnormal values and does not generate the problem of serious distorted data distribution due to the objects with the maximum values, and the k-means algorithm is the object which is positioned at the center in the cluster and does not take the average value in the cluster as the reference point.
Specifically, the step of constructing the life cycle curve model in the step 5 is as follows: from establishment of an enterprise to the present, model construction is carried out on the related application amount and the related patent application amount of a certain technology, wherein time is an X axis, the number of patent applications is a Y axis, a two-dimensional S curve is drawn, logistic regression fitting is carried out on the two-dimensional S curve, and the life cycle of the technology is obtained in a quantitative and qualitative combination mode according to preset technology growth rate and technology maturity coefficient evaluation indexes.
Furthermore, the life cycle curve model is constructed by adopting the S curve, the general rules of the patent application amount and the practice of the enterprise in the period from establishment of the enterprise to the current time can be visually displayed, four stages of the technical life cycle are obtained by analyzing the relationship between the patent application amount and the practice, and the patent of each stage can generate different characteristics. At the initial stage of establishment of an enterprise, the technology is in the germination stage at the moment, the number of patents is small, and only a few technologies are researched and developed in the market and appear as basic patents; with the short-term development of enterprises, the technology is in the growth period at the moment, the research and development investment of the technology is increased, and the number of patents is increased sharply; with the long-term slow development of enterprises, the technology is in the mature stage at this time, and the number of patents tends to be slow; with the development of new business by enterprises, the prior art is in the decline period at this time, and the number of patents is increased negatively.
Further, the present invention relates to a technology growth rate which is a ratio of a patent application amount of a certain technology to an application amount of a certain past time.
Furthermore, the technical maturity coefficient related to the invention is the ratio of the patent application amount of a certain technical patent to the patent application amount of the technical field.
Furthermore, the invention also combines the clustering result with the life cycle curve model to obtain the intellectual property data analysis result, thereby further improving the reliability and the practicability of the patent layout.
Finally, the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all of them should be covered in the claims of the present invention.
Claims (5)
1. An intellectual property data analysis method based on technical positions is characterized by comprising the following steps:
step 1: the method comprises the steps of butting with a patent database of the national intellectual property office through an API (application programming interface), inputting a name of a user enterprise to obtain a patent of the user enterprise and storing the patent in a cloud database;
step 2: indexing the technical and efficacy indexes of the specifications in the patents of the user enterprises to obtain a technical feature and achieved technical and efficacy set, and constructing a technical function matrix;
and step 3: calculating the technical similarity between the patent and the specification by using a cosine included angle method;
and 4, step 4: performing patent clustering analysis to obtain a clustering result;
and 5: constructing a life cycle curve model;
step 6: and combining the clustering result with the life cycle curve model to obtain an intellectual property data analysis result.
2. The method for analyzing intellectual property data based on technical position as claimed in claim 1, wherein the specific step of indexing the claim in the user enterprise patent in step 2 is as follows: performing NLP natural language processing on the specifications in the patents of the user enterprises, performing word segmentation, word stop removal and word frequency statistics, extracting technical keywords and achieved technical effects, ranking according to word frequency, and converting into a technical function matrix Csj＝{(T11，E11)，(T12，E12)，...，(Tsj，Esj)，...，(Twn，Ewn) A, (j ═ 1, 2, 3.., n), (s ═ 1, 2, 3.., w), where T is presentsjShowing the technical characteristics in patent s, EsjIndicating the generation of corresponding technical features in patent sHas the effects of relieving fatigue.
3. The method as claimed in claim 1, wherein the cosine angle in step 3 is expressed by the following formula
Wherein Ai is composed of { kCa1，kCa2，kCa3，...，kCaj，...，kCanComposed of { kC }, Bib1，kCb2，kCb3，...，kCbj，...，kCbnIn which, kCajIs (T)aj，Eaj) The number of occurrences in patent a, (a ═ 1, 2, 3,. eta., w), kCbjIs (T)bj，Ebj) In the number of occurrences in patent b, (b ═ 1, 2, 3., w), when similarity is 0, i.e., the degree of similarity between patent a and patent b is maximum, and when similarity becomes gradually larger, i.e., the degree of similarity between patent a and patent b becomes gradually smaller, the previous steps are repeated and the similarity values are converted into a similarity matrix.
4. The method for analyzing intellectual property data based on technical position as claimed in claim 1, wherein the proprietary clustering analysis method in step 4 is k-medoids algorithm, comprising the following steps:
(41) classifying the patents of the user enterprises according to the IPC classification number to obtain q total classes;
(42) randomly selecting r classes from q total classes as reference points Zt(t＝1，2，3，...，r)；
(43) Distributing the remaining q-r technical subjects to each cluster class according to the principle closest to medoids;
(44) for all classes except the corresponding reference point in the t-th cluster class, sequentially calculating the value of the criterion function when the class is a new reference point, traversing all possibilities, and selecting the corresponding class when the criterion function is minimum as a new cluster class;
(45) repeating the processes (43) and (44) until all classes are not changed or the set maximum iteration number is reached;
(46) finally, r cluster classes are determined.
5. The method for analyzing intellectual property data based on technical position as claimed in claim 1, wherein the step of constructing the life cycle curve model in the step 5 is as follows: from establishment of an enterprise to the present, model construction is carried out on the related application amount and the related patent application amount of a certain technology, wherein time is an X axis, the number of patent applications is a Y axis, a two-dimensional S curve is drawn, logistic regression fitting is carried out on the two-dimensional S curve, and the life cycle of the technology is obtained in a quantitative and qualitative combination mode according to preset technology growth rate and technology maturity coefficient evaluation indexes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110525628.3A CN113326372A (en) | 2021-05-13 | 2021-05-13 | Intellectual property data analysis method based on technical position |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110525628.3A CN113326372A (en) | 2021-05-13 | 2021-05-13 | Intellectual property data analysis method based on technical position |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113326372A true CN113326372A (en) | 2021-08-31 |
Family
ID=77415779
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110525628.3A Pending CN113326372A (en) | 2021-05-13 | 2021-05-13 | Intellectual property data analysis method based on technical position |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113326372A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114781553A (en) * | 2022-06-20 | 2022-07-22 | 浙江大学滨江研究院 | Unsupervised patent clustering method based on parallel multi-graph convolution neural network |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101989267A (en) * | 2009-07-30 | 2011-03-23 | 上海汉光知识产权数据科技有限公司 | List analysis system and method for patent analysis |
| CN105260958A (en) * | 2015-11-20 | 2016-01-20 | 上海熠派信息科技有限公司 | Intellectual property right one-stop service system |
| CN108805458A (en) * | 2018-06-20 | 2018-11-13 | 国网电力科学研究院武汉南瑞有限责任公司 | A kind of enterprise technology Competitiveness Assessment method and device |
| CN110348133A (en) * | 2019-07-15 | 2019-10-18 | 西南交通大学 | A kind of bullet train three-dimensional objects structure technology effect figure building system and method |
| CN110956446A (en) * | 2019-11-27 | 2020-04-03 | 苏州骐越信息科技有限公司 | Intellectual property one-stop service system |
-
2021
- 2021-05-13 CN CN202110525628.3A patent/CN113326372A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101989267A (en) * | 2009-07-30 | 2011-03-23 | 上海汉光知识产权数据科技有限公司 | List analysis system and method for patent analysis |
| CN105260958A (en) * | 2015-11-20 | 2016-01-20 | 上海熠派信息科技有限公司 | Intellectual property right one-stop service system |
| CN108805458A (en) * | 2018-06-20 | 2018-11-13 | 国网电力科学研究院武汉南瑞有限责任公司 | A kind of enterprise technology Competitiveness Assessment method and device |
| CN110348133A (en) * | 2019-07-15 | 2019-10-18 | 西南交通大学 | A kind of bullet train three-dimensional objects structure technology effect figure building system and method |
| CN110956446A (en) * | 2019-11-27 | 2020-04-03 | 苏州骐越信息科技有限公司 | Intellectual property one-stop service system |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114781553A (en) * | 2022-06-20 | 2022-07-22 | 浙江大学滨江研究院 | Unsupervised patent clustering method based on parallel multi-graph convolution neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107633007B (en) | Commodity comment data tagging system and method based on hierarchical AP clustering | |
| US7299247B2 (en) | Construction of trainable semantic vectors and clustering, classification, and searching using trainable semantic vectors | |
| WO2020199591A1 (en) | Text categorization model training method, apparatus, computer device, and storage medium | |
| CN111104466B (en) | Method for quickly classifying massive database tables | |
| US8126826B2 (en) | Method and system for active learning screening process with dynamic information modeling | |
| Liu et al. | Combining enterprise knowledge graph and news sentiment analysis for stock price prediction | |
| US20060242190A1 (en) | Latent semantic taxonomy generation | |
| CN111062757A (en) | Information recommendation method and system based on multi-path optimization matching | |
| CN112632228A (en) | Text mining-based auxiliary bid evaluation method and system | |
| CN104077407B (en) | A kind of intelligent data search system and method | |
| CN114119057B (en) | User portrait model construction system | |
| CN111489201A (en) | Method, device and storage medium for analyzing customer value | |
| CN111368096A (en) | Knowledge graph-based information analysis method, device, equipment and storage medium | |
| CN114119058A (en) | User portrait model construction method and device and storage medium | |
| Bhardwaj et al. | Review of text mining techniques | |
| CN116431931A (en) | Real-time incremental data statistical analysis method | |
| Tavakoli et al. | Clustering time series data through autoencoder-based deep learning models | |
| CN113570380A (en) | Service complaint processing method, device and equipment based on semantic analysis and computer readable storage medium | |
| CN113326372A (en) | Intellectual property data analysis method based on technical position | |
| CN116910599A (en) | Data clustering method, system, electronic equipment and storage medium | |
| Lo et al. | An emperical study on application of big data analytics to automate service desk business process | |
| CN110414819B (en) | Work order scoring method | |
| CN112818215A (en) | Product data processing method, device, equipment and storage medium | |
| CN112100370B (en) | Picture-trial expert combination recommendation method based on text volume and similarity algorithm | |
| CN115168408B (en) | Query optimization method, device, equipment and storage medium based on reinforcement learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210831 |
|
| RJ01 | Rejection of invention patent application after publication |