CN110727801A - Ontology-based fuzzy evaluation search engine advertisement optimization method - Google Patents
Ontology-based fuzzy evaluation search engine advertisement optimization method Download PDFInfo
- Publication number
- CN110727801A CN110727801A CN201910268626.3A CN201910268626A CN110727801A CN 110727801 A CN110727801 A CN 110727801A CN 201910268626 A CN201910268626 A CN 201910268626A CN 110727801 A CN110727801 A CN 110727801A
- Authority
- CN
- China
- Prior art keywords
- advertisement
- search engine
- ontology
- evaluation
- optimization
- 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
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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Computational Linguistics (AREA)
- Game Theory and Decision Science (AREA)
- Animal Behavior & Ethology (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to the field of data processing, in particular to a fuzzy evaluation search engine advertisement optimization method based on ontology, which comprises the following steps: s1: constructing a search engine advertisement semantic model of the ontology, describing search engine advertisement field knowledge at a semantic level and multiplexing and integrating the existing ontology model to realize information exchange and sharing; s2: based on a multi-attribute entropy weight fuzzy comprehensive evaluation algorithm, calculating attribute weights by using information entropy, constructing a membership matrix by using triangular fuzzy numbers, constructing a multi-attribute comment evaluation coordinate system according to an optimization target and an optimization rule, and calculating by using a fuzzy operator to obtain a final optimized subset. The technology can automatically realize the early work (rough optimization) of the search engine advertisement optimization, and complete the more advanced deep optimization to a certain extent, thereby effectively saving the labor cost for enterprises and improving the investment return rate of the search engine popularization.
Description
Technical Field
The invention belongs to the field of data processing, and particularly relates to a fuzzy evaluation search engine advertisement optimization method based on an ontology.
Background
The Search Engine advertisement has the characteristics of strong pertinence, low cost, high flexibility, easy release and the like compared with the traditional advertisement media, the advertiser mainly focuses on three aspects in the whole Search Engine advertisement putting process, namely ① creating advertisement ② monitoring data ③ to optimize the advertisement, and Search Engine advertisement Optimization (Search Engine Optimization) becomes the focus of the current advertiser in order to make the advertisement putting more effective.
Currently, relevant research focuses on both model-based and rule-based optimization. However, the advertisement promotion effect is a result of multi-attribute comprehensive influence, and the ideal advertisement delivery effect cannot be achieved by predicting a single attribute before advertisement delivery by adopting model-based optimization. Rule-based optimization is actually a process of comprehensively evaluating advertisement attributes according to optimization objectives and related optimization rules, screening out an optimization subset conforming to the rules, and implementing an optimization strategy. In the process, advertisement subsets obtained by comprehensive evaluation and screening are used as implementation objects of the optimization strategy, and the quality of the advertisement subsets directly influences the optimization effect. Today, an advertiser may have tens of thousands to hundreds of thousands of advertisements, and how to accurately locate a proper advertisement in a large amount of advertisement data to implement an optimization strategy has become a great challenge. Traditional SEO relies entirely on manual analysis, screening of data. The manual optimization method mainly has the following defects:
1) the manual analysis process often depends on field experience, and the subjectivity is strong.
2) Optimization goals typically involve multiple advertising attributes, and manual optimization approaches have difficulty in making accurate comprehensive assessments of the numerous attributes that have complex associations.
3) The fuzzy-described optimization rules have difficulty in establishing an accurate mapping relationship with the numerically-described advertisement attributes, which further increases the difficulty for the site-optimized subsets.
Disclosure of Invention
Aiming at the problems, the invention provides a search engine advertisement semantic model based on an ontology, aiming at describing search engine advertisement field knowledge and multiplexing and integrating the existing ontology model at a semantic level so as to realize information exchange and sharing; on the other hand, a search engine advertisement optimization strategy based on multi-attribute entropy weight fuzzy comprehensive evaluation is provided, the algorithm uses information entropy to calculate attribute weight, a triangular fuzzy number is used for constructing a membership matrix, a multi-attribute comment evaluation coordinate system is constructed according to an optimization target and an optimization rule, and a final optimization subset is obtained through fuzzy operator calculation, and the specific method comprises the following steps:
a fuzzy evaluation search engine advertisement optimization method based on ontology is characterized in that: the method comprises the following steps:
s1: constructing a search engine advertisement semantic model of the ontology, describing search engine advertisement field knowledge at a semantic level and multiplexing and integrating the existing ontology model, and realizing information exchange and sharing:
s2: based on a multi-attribute entropy weight fuzzy comprehensive evaluation algorithm, calculating attribute weights by using information entropy, constructing a membership matrix by using triangular fuzzy numbers, constructing a multi-attribute comment evaluation coordinate system according to an optimization target and an optimization rule, and calculating by using a fuzzy operator to obtain a final optimized subset.
Further, the step S1 includes the following steps:
s11: formalized definition of search engine advertisements, wherein the organization structure of the search engine advertisements is an advertisement instance set consisting of a plurality of advertisement instances; the advertisement instance is described by an advertisement attribute set, an advertisement data set and an advertisement behavior set triple;
s12: designing an organization structure of a search engine advertisement semantic model, and abstracting the semantic model into three top-level objects including a core attribute, a core object and a core behavior according to triple definitions of advertisement instances; meanwhile, in order to enhance the semantic representation capability of the model, Adfeature is extended to be used as a subclass of a core object to describe advertisement characteristics.
S13: constructing a search and talk engine advertisement semantic model of the ontology: the ontology construction method based on the spiral model is adopted, the semantic model of the search engine ontology is defined as a quintuple O (C, P, I, H, A), C represents a concept set, P represents an attribute set, I represents an instance set, H describes the hierarchical relationship of C, P, I, and A represents an inference rule set; s12, mapping the class, relation and organization structure in the semantic model into a concept set C, an attribute P, a hierarchical relation H and an inference rule A in the ontology model, and then filling an instance set I into the ontology semantic model.
Further, the step S2 includes the following steps:
s21: establishing a multi-factor comment index system:
(1) mapping the SEOBehavior class into an optimization target set T ═ T according to an optimization behavior knowledge set defined in a search engine advertisement body modeliI ∈ N }, where T isiThe ith optimization behavior object is in the SEAO model, and the ith optimization target is represented in the optimization target set T;
(2) optimizing a target TiThe technique defines the evaluation factor set U ═ U as being associated with SEORule by the hasSEOR attribute, while the hasDelatedData attribute describes the advertising data associated with SEORule1,…,Uj,…UqIn which U isjTo optimize the target TiThe corresponding j-th correlation factor;
(3) the DataType type defined in the SEAO model describes the data type of the search engine advertising effect data, and the data type is numerical type, composite type and nominal type respectively;
(4) the nominal data value field has an enumeration property, and a comment set V is defined as { Ture, False }; the set of numeric and coincidence data comments is defined as V ═ { High, Middle, Low };
(5) defining a multi-factor comment evaluation index system S ═<U+,T+>Wherein U is+Set of assessment factors, T, representing semantic description of the attached comment+Representing the optimization target with comment semantic description, then optimizing the target TiMulti-factor comment evaluation index system for evaluation factor set U
S22: establishing an evaluation matrix:
(1) with m advertisements, the evaluation matrix is initialized to
xhjA real value representing the h advertisement on the i effect attribute;
(2) normalizing the initialized matrix X to obtain a standard matrix omega
h∈[1,m],m∈N*,j∈[1,k],k∈N*
S23: determining a weight vector: calculating objective weight according to each index by using an information entropy tool; using an expert investigation method to count subjective weights of all indexes; nominal index and T+The index does not participate in the weight calculation, and defines the weight vector of the factor set
S24: fuzzy comprehensive evaluation: the method comprises the following specific steps:
(1) initializing membership matrix
The technique uses Boolean functions and triangular fuzzy to calculate fuzzy membership, which is defined as follows:
wherein z is a sliding constant, q1Is the lower quartile, q2Is a median, q3Is the upper quartile, q4=q3+1.5IQR, IQR is a quartering distance;
(2) will (x)hj) Substituting the formula (7) to calculate an evaluation membership degree matrix R,
(3) calculating a weighting matrix
Firstly, according to the membership degree matrix R, eliminating the samples with the membership degree of 0 in the KPI index rows and the nominal index rows. Then calculating a weighting matrix
(4) Combining the evaluation results
To weighting matrixCarrying out statistical summation to obtain a multi-factor comment evaluation coordinate system S of each advertisement instanceiEvaluation result vector A of
And sorting the result vector A, and selecting to obtain a final optimized subset by referring to an optimization strength threshold delta given by an expert.
Further, the ontology construction method based on the spiral model comprises the five steps of ① mapping objects in a knowledge set to be a concept set C of the ontology model according to the extracted hierarchical structure, ② constructing a property set P according to the extracted association relation, ③ extracting advertisement examples in a data source to establish an example set, ④ constructing an inference rule set according to the inference relation among the concept set, the property set and the example set, ⑤ evaluating the model according to an expert evaluation standard, and repeating the steps of 1-4 until a complete ontology model is obtained.
Further, w in the step S23jThe specific calculation steps are as follows:
(1) the entropy of the j index is calculated from the normalization matrix Ω:
(2) Calculate the difference of the jth index:
dj=1-ej
(3) determining the entropy weight of the index j:
(4) determining the weight of the index j:
wj=λjα+βj(1-α),α∈[0,1]
wherein alpha is a regulatory factor, betajSubjective weights given to experts.
The invention provides a search engine advertisement semantic model based on an ontology, which aims to describe search engine advertisement field knowledge and multiplex and integrate the existing ontology model at a semantic level so as to realize information exchange and sharing; and on the other hand, a search engine advertisement optimization strategy based on multi-attribute entropy weight fuzzy comprehensive evaluation is provided, the algorithm uses the information entropy to calculate the attribute weight, a triangular fuzzy number is used for constructing a membership matrix, a multi-attribute comment evaluation coordinate system is constructed according to an optimization target and an optimization rule, and a final optimization subset is obtained through fuzzy operator calculation. The technology can automatically realize the early work (rough optimization) of the search engine advertisement optimization, and complete the more advanced deep optimization to a certain extent, thereby effectively saving the labor cost for enterprises and improving the investment return rate of the search engine popularization.
Drawings
FIG. 1 is an organizational structure of a search engine advertisement as defined by the present invention.
FIG. 2 is an organizational structure of a search engine advertising semantic model designed according to the present invention.
Detailed Description
The invention is described in further detail below with reference to the drawings.
The invention provides a fuzzy evaluation search engine advertisement optimization method based on a body, which comprises two technical cores:
(1) providing a search engine advertisement semantic model based on an ontology, aiming at describing search engine advertisement field knowledge at a semantic level and multiplexing and integrating the existing ontology model so as to realize information exchange and sharing;
(2) the algorithm uses the information entropy to calculate the attribute weight, utilizes the triangular fuzzy number to construct a membership matrix, constructs a multi-attribute comment evaluation coordinate system according to an optimization target and an optimization rule, and obtains a final optimization subset through fuzzy operator calculation.
Through the mutual combination of the two, the manual optimization process can be effectively simulated, the defects of the traditional manual optimization are avoided, and the automatic processing of the search engine advertisement optimization is realized. The specific implementation steps are as follows:
1. ontology-based search engine advertisement semantic model
The method for constructing the search engine advertisement semantic model of the ontology mainly comprises the following steps:
(11) the organization structure of the search engine advertisement is defined as shown in figure one. The formalization of a search engine advertisement is defined as follows:
definition 1: SEA ═ Adi|Adi=<P,D,H>,1≤i≤n,n∈N+}
Definition 2: p ═ Domain priority, AdStructure }
Definition 3: d ═ Di|di=<AD,BasicInfo,ED,TD,DT>,1≤i≤n,n∈N+}
Definition 4: h ═ CreateBehavior, SEOBehavior }
Where SEA is a set of advertisement instances, ad instances Adi are described by a triple of a set of ad attributes P, a set of ad data D, and a set of ad behaviors H. The advertisement attribute set P is composed of a domain concept knowledge set domainpriority and an advertisement structure knowledge set adstrucurepriority, as defined in definition 2; definition 3 gives a formal description of an advertisement data set D, wherein AD is attribute data, Basicinfo is basic information data, ED is effect data, TD is flow data, and DT is a data type; h is the set of behaviors imposed on top of the instance of the ad, including the set of ad creation behaviors CreatBehavior and the ad optimization behavior SEOBehavior, given by definition 4.
(12) The organizational structure of the semantic model of the search engine advertisement is designed, as shown in figure two. According to the triple definition of the advertisement, the semantic model is abstracted into three top-level objects of a core attribute, a core object and a core behavior. Meanwhile, in order to enhance the semantic representation capability of the model, Adfeature is expanded to be used as a subclass of a core object for describing advertisement characteristics.
(13) And constructing an ontology semantic model. And adopting a body construction method based on a spiral model. The search engine ontology model is defined as five-tuple O ═ C, (P, I, H, a), C represents a concept set, P is an attribute set, I is an instance set, H describes the hierarchical relationship of C, P, I, and a is an inference rule set. The class, relation and organizational structure in the semantic model are mapped into a concept set C, an attribute P, a hierarchical relation H and an inference rule A in the ontology model. Instance set A is then populated into the onto-model. The specific model construction steps are as follows:
a) defining the definition category of the ontology, and determining the category of the ontology: considering first the possibility of reusing existing ontologies, no more sophisticated search engine advertising ontology has been proposed. The search engine advertisement ontology should therefore have the ability to describe both search engine advertisement domain knowledge and to integrate existing ontologies as a top-level ontology.
b) Collecting a data source and extracting a knowledge set: and extracting to obtain a knowledge set according to a representation frame defined by the search engine advertisement semantic model.
c) The ontology model building method based on the spiral model mainly comprises the five steps of ① mapping objects in a knowledge set into a concept set C of the ontology model according to an extracted hierarchical structure, ② building a property set P according to an extracted incidence relation, ③ extracting advertisement examples in a data source to build an example set, ④ building an inference rule set according to an inference relation among the concept set, the property set and the example set, ⑤ evaluating the model according to an expert evaluation standard, and repeating 1-4 steps until a complete ontology model is obtained
2. Multi-attribute entropy weight fuzzy comprehensive evaluation algorithm
The algorithm mainly comprises the following steps:
(21) establishing a multi-factor comment coordinate system: mapping the SEOBehavior class into an optimization target set T ═ T according to an optimization behavior knowledge set defined in a search engine advertisement body modeliL 1 ∈ N }, where TiIn the SEAO model is the ith optimization behavior object, which represents the ith optimization objective in the optimization objective set T. Optimizing a target TiThe technique defines the evaluation factor set U ═ U as being associated with SEORule by the hasSEOR attribute, while the hasDelatedData attribute describes the advertising data associated with SEORule1,…,Uj,…UqIn which U isjTo optimize the target TiThe corresponding j-th correlation factor. The DataType class defined in the SEAO model describes the data types of the search engine advertisement effectiveness data, numeric, composite, and nominal, respectively. The nominal data value field has an enumeration property, and a comment set V is defined as { Ture, False }; the numeric and coincidence data panel is defined as V ═ High, Middle, Low. Defining a multi-factor comment evaluation index system S ═<U+,T+>Wherein U is+Set of assessment factors, T, representing semantic description of the attached comment+Representing the optimization target with comment semantic description, then optimizing the target TiMulti-factor comment evaluation index system for evaluation factor set UFor example: comment evaluation index system<High display, low consumption, medium impact amount, medium conversion amount and medium conversion rate>Wherein U is+=<High exhibition, low consumption, middle striking amount and transfer amount>,T+=<Medium conversion rate>。
(22) Establishing an evaluation matrix:
with m advertisements, the evaluation matrix is initialized to
xhjRepresenting the true value of the h ad on the i-th performance attribute. Normalizing the initialized matrix X to obtain a standard matrix omega
(23) Determining a weight vector
The entropy characterizes the purity of any sample set. The larger the information entropy of the evaluation factor is, the larger the variation degree of the value is, the more information is provided in the comprehensive evaluation process, and the larger the weight of the index is. Therefore, the objective weight of each index is calculated by using the information entropy tool. On the other hand, subjective weights of the respective indices are counted using an expert survey method. Due to the nominal index and T+The index has semantic duality, so the index does not participate in weight calculation. Defining a factor set weight vectorwjThe specific calculation steps are as follows:
a) the entropy of the j index is calculated from the normalization matrix Ω:
b) Calculate the difference of the jth index:
dj=1-ej
c) determining the entropy weight of the index j:
d) determining the weight of the index j:
wj=λjα+βj(1-α),α∈[0,1]
wherein alpha is a regulatory factor, betajSubjective weights given to experts.
(24) Fuzzy comprehensive evaluation
a) Initializing membership matrix
The technique uses Boolean functions and triangular fuzzy to calculate fuzzy membership, which is defined as follows:
wherein q is1Is the lower quartile, q2Is a median, q3Is the upper quartile, q4=q3+1.5IQR, IQR is a quartering distance.
Will (x)hj) Substituting the formula (7) to calculate an evaluation membership degree matrix R,
b) calculating a weighting matrix
Firstly, according to the membership degree matrix R, eliminating the samples with the membership degree of 0 in the KPI index rows and the nominal index rows. Then calculating a weighting matrix
c) Combining the evaluation results
To weighting matrixCarrying out statistical summation to obtain a multi-factor comment evaluation coordinate system S of each advertisement instanceiEvaluation result vector A of
And sorting the result vector A, and selecting to obtain a final optimized subset by referring to an optimization strength threshold delta given by an expert.
Claims (5)
1. A fuzzy evaluation search engine advertisement optimization method based on ontology is characterized in that: the method comprises the following steps:
s1: constructing a search engine advertisement semantic model of the ontology, describing search engine advertisement field knowledge at a semantic level and multiplexing and integrating the existing ontology model to realize information exchange and sharing;
s2: based on a multi-attribute entropy weight fuzzy comprehensive evaluation algorithm, calculating attribute weights by using information entropy, constructing a membership matrix by using triangular fuzzy numbers, constructing a multi-attribute comment evaluation coordinate system according to an optimization target and an optimization rule, and calculating by using a fuzzy operator to obtain a final optimized subset.
2. The ontology-based fuzzy evaluation search engine advertisement optimization method of claim 1, wherein: the step S1 includes the following steps:
s11: formalized definition of search engine advertisements, wherein the organization structure of the search engine advertisements is an advertisement instance set consisting of a plurality of advertisement instances; the advertisement instance is described by an advertisement attribute set, an advertisement data set and an advertisement behavior set triple;
s12: designing an organization structure of a search engine advertisement semantic model, and abstracting the semantic model into three top-level objects including a core attribute, a core object and a core behavior according to triple definitions of advertisement instances; meanwhile, in order to enhance the semantic representation capability of the model, Adfeature is extended to be used as a subclass of a core object to describe advertisement characteristics.
S13: constructing a search and talk engine advertisement semantic model of the ontology: the ontology construction method based on the spiral model is adopted, the semantic model of the search engine ontology is defined as a quintuple O (C, P, I, H, A), C represents a concept set, P represents an attribute set, I represents an instance set, H describes the hierarchical relationship of C, P, I, and A represents an inference rule set; s12, mapping the class, relation and organization structure in the semantic model into a concept set C, an attribute P, a hierarchical relation H and an inference rule A in the ontology model, and then filling an instance set I into the ontology semantic model.
3. The ontology-based fuzzy evaluation search engine advertisement optimization method of claim 1, wherein: the step S2 includes the following steps:
s21: establishing a multi-factor comment index system:
(1) mapping the SEOBehavior class into an optimization target set T ═ T according to an optimization behavior knowledge set defined in a search engine advertisement body modeli|i∈N*Where T isiIn the SEAO model is the ith optimizationBehavior objects, representing the ith optimization objective in the optimization objective set T;
(2) optimizing a target TiThe technique defines the evaluation factor set U ═ U as being associated with SEORule by the hasSEOR attribute, while the hasDelatedData attribute describes the advertising data associated with SEORule1,…,Uj,…UqIn which U isjTo optimize the objective, TiThe corresponding j-th correlation factor;
(3) the DataType type defined in the SEAO model describes the data type of the search engine advertising effect data, and the data type is numerical type, composite type and nominal type respectively;
(4) the nominal data value field has an enumeration property, and a comment set V is defined as { Ture, False }; the set of numeric and coincidence data comments is defined as V ═ { High, Middle, Low };
(5) defining a multi-factor comment evaluation index system S ═<U+,T+>Wherein U is+Set of assessment factors, T, representing semantic description of the attached comment+Representing the optimization target with comment semantic description, then optimizing the target TiMulti-factor comment evaluation index system for evaluation factor set U
S22: establishing an evaluation matrix:
(1) with m advertisements, the evaluation matrix is initialized to
h∈[1,m],m∈N*,j∈[1,k],k∈N*xhjRepresenting the true value of the h advertisement on the j effect attribute;
(2) normalizing the initialized matrix X to obtain a standard matrix omega
S23: determining a weight vector: calculating objective weight according to each index by using an information entropy tool; using an expert investigation method to count subjective weights of all indexes; nominal index and T+The index does not participate in the weight calculation, and defines the weight vector of the factor set
S24: fuzzy comprehensive evaluation: the method comprises the following specific steps:
(1) initializing membership matrix
The technique uses Boolean functions and triangular fuzzy to calculate fuzzy membership, which is defined as follows:
wherein z is a sliding constant, q1Is the lower quartile, q2Is a median, q3Is the upper quartile, q4=q3+1.5IQR, IQR is a quartering distance;
(2) will (x)hj) Evaluation by substituting formula (7)The matrix of degrees of membership R is,
(3) calculating a weighting matrix
Firstly, according to the membership degree matrix R, eliminating the samples with the membership degree of 0 in the KPI index rows and the nominal index rows. Then calculating a weighting matrix
(4) Combining the evaluation results
To weighting matrixCarrying out statistical summation to obtain a multi-factor comment evaluation coordinate system S of each advertisement instanceiEvaluation result vector A of
And sorting the result vector A, and selecting to obtain a final optimized subset by referring to an optimization strength threshold delta given by an expert.
4. The ontology-based fuzzy evaluation search engine advertisement optimization method of claim 2, wherein the ontology construction method based on the spiral model is divided into five steps, ① mapping objects in a knowledge set to be a concept set C of the ontology model according to the extracted hierarchical structure, ② constructing an attribute set P according to the extracted association relationship, ③ extracting advertisement examples in a data source to establish an example set, ④ constructing an inference rule set according to the inference relationship among the concept set, the attribute set and the example set, ⑤ evaluating the model according to an expert evaluation standard, and repeating the steps 1-4 until a complete ontology model is obtained.
5. The ontology-based fuzzy evaluation search engine advertisement optimization method of claim 3, wherein: w in said step S23jThe specific calculation steps are as follows:
(1) the entropy of the j index is calculated from the normalization matrix Ω:
(2) Calculate the difference of the jth index:
dj=1-ej
(3) determining the entropy weight of the index j:
(4) determining the weight of the index j:
wj=λjα+βj(1-α),α∈[0,1]
wherein alpha is a regulatory factor, betajSubjective weights given to experts.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910268626.3A CN110727801A (en) | 2019-04-04 | 2019-04-04 | Ontology-based fuzzy evaluation search engine advertisement optimization method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910268626.3A CN110727801A (en) | 2019-04-04 | 2019-04-04 | Ontology-based fuzzy evaluation search engine advertisement optimization method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110727801A true CN110727801A (en) | 2020-01-24 |
Family
ID=69217668
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910268626.3A Pending CN110727801A (en) | 2019-04-04 | 2019-04-04 | Ontology-based fuzzy evaluation search engine advertisement optimization method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110727801A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112417160A (en) * | 2020-11-11 | 2021-02-26 | 北京机电工程总体设计部 | Semantic knowledge association method oriented to index system |
CN117950787A (en) * | 2024-03-22 | 2024-04-30 | 成都赛力斯科技有限公司 | Advertisement display method and device, electronic equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20110061451A (en) * | 2009-12-01 | 2011-06-09 | 하현숙 | Concept image data retrieval method and device based on linguistic analysis, fuzzy color information and semantic web |
KR20140026932A (en) * | 2012-08-24 | 2014-03-06 | (주)에이엔비 | System and method providing a suited shopping information by analyzing the propensity of an user |
CN105868115A (en) * | 2016-04-05 | 2016-08-17 | 工业和信息化部电子第五研究所 | Building method and system for software test model of software intensive system |
CN108256761A (en) * | 2018-01-15 | 2018-07-06 | 中国人民解放军陆军装甲兵学院 | Overhead weapon station grey fuzzy comprehensive evaluation method |
-
2019
- 2019-04-04 CN CN201910268626.3A patent/CN110727801A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20110061451A (en) * | 2009-12-01 | 2011-06-09 | 하현숙 | Concept image data retrieval method and device based on linguistic analysis, fuzzy color information and semantic web |
KR20140026932A (en) * | 2012-08-24 | 2014-03-06 | (주)에이엔비 | System and method providing a suited shopping information by analyzing the propensity of an user |
CN105868115A (en) * | 2016-04-05 | 2016-08-17 | 工业和信息化部电子第五研究所 | Building method and system for software test model of software intensive system |
CN108256761A (en) * | 2018-01-15 | 2018-07-06 | 中国人民解放军陆军装甲兵学院 | Overhead weapon station grey fuzzy comprehensive evaluation method |
Non-Patent Citations (7)
Title |
---|
QING YANG ETC.: "Fuzzy ontology generation model using fuzzy clustering for learning evaluation", 《2009 IEEE INTERNATIONAL CONFERENCE ON GRANULAR COMPUTING》 * |
YU YANG-XIN: "Fuzzy Information Retrieval Model Based on Multiple Related Ontologies", 《COMPUTER ENGINEERING》 * |
但斌等: "基于本体映射面向模糊客户需求的产品配置研究", 《计算机集成制造***》 * |
刘昕民 等: "基于模糊领域本体的专家遴选服务研究", 《北京理工大学学报》 * |
张博 等: "面向多视点的商品本体建模", 《情报杂志》 * |
张祥: "基于本体推理及模糊评价的在线学习***设计与实现", 《中国优秀硕士学位论文全文数据库(信息科技辑)》 * |
李德仁等: "基于时空模糊本体的交通领域知识建模", 《武汉大学学报(信息科学版)》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112417160A (en) * | 2020-11-11 | 2021-02-26 | 北京机电工程总体设计部 | Semantic knowledge association method oriented to index system |
CN112417160B (en) * | 2020-11-11 | 2023-07-25 | 北京机电工程总体设计部 | Semantic knowledge association method oriented to index system |
CN117950787A (en) * | 2024-03-22 | 2024-04-30 | 成都赛力斯科技有限公司 | Advertisement display method and device, electronic equipment and storage medium |
CN117950787B (en) * | 2024-03-22 | 2024-05-31 | 成都赛力斯科技有限公司 | Advertisement display method and device, electronic equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liao et al. | A survey of decision-making methods with probabilistic linguistic information: bibliometrics, preliminaries, methodologies, applications and future directions | |
Wang et al. | Distributed energy system for sustainability transition: A comprehensive assessment under uncertainties based on interval multi-criteria decision making method by coupling interval DEMATEL and interval VIKOR | |
Wu et al. | Evaluation the drivers of green supply chain management practices in uncertainty | |
Kizielewicz et al. | Comparison of Fuzzy TOPSIS, Fuzzy VIKOR, Fuzzy WASPAS and Fuzzy MMOORA methods in the housing selection problem | |
CN105117422A (en) | Intelligent social network recommender system | |
CN103632203A (en) | Distribution network power supply area division method based on comprehensive evaluation | |
CN110990718B (en) | Social network model building module of company image lifting system | |
Liu et al. | Defective alternatives detection-based multi-attribute intuitionistic fuzzy large-scale decision making model | |
CN109934469A (en) | Based on the heterologous power failure susceptibility method for early warning and device for intersecting regression analysis | |
CN114529038A (en) | Intelligent matching business recruitment strategy system and method based on enterprise demands | |
CN110727801A (en) | Ontology-based fuzzy evaluation search engine advertisement optimization method | |
CN116128544A (en) | Active auditing method and system for electric power marketing abnormal business data | |
CN114912772A (en) | Urban right transparency differential evaluation system matching method and system based on urban economic classification analysis | |
Zuo et al. | A property perceived service quality evaluation method for public buildings based on multisource heterogeneous information fusion | |
Cong et al. | New smart city clusters’ construction level evaluation under economic circles: The case of Shandong, China | |
Inyang et al. | Visual association analytics approach to predictive modelling of students’ academic performance | |
Zhou et al. | A statistical approach to large-scale group decision making with incomplete hesitant fuzzy linguistic information by incorporating risk attitude | |
CN113837913B (en) | Method and device for determining key threshold value of rural cultivated land resource bearing capacity | |
Kvyatkovskaya et al. | Methodology of a support of making management decisions for poorly structured problems | |
Wang | [Retracted] Construction and Effect Analysis of College Students’ Physical Education Teaching Mode Based on Data Mining Algorithm | |
Karimi et al. | Analyzing the results of buildings energy audit by using grey incidence analysis | |
Khomonenko et al. | Approach to processing of data from social networks for detecting public opinion on quality of educational services | |
Zheng et al. | Demand-driven NEV supplier selection: An integrated method based on ontology–QFD–CBR | |
CN110598973A (en) | IAP-based risk evaluation method for authentication process of green furniture product | |
Jiang et al. | Community Detection using Closeness Similarity based on Common Neighbor Node Clustering Entropy. |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200124 |