CN115098766B - Bidding information recommendation method and system for electronic bidding transaction platform - Google Patents
Bidding information recommendation method and system for electronic bidding transaction platform Download PDFInfo
- Publication number
- CN115098766B CN115098766B CN202210590392.6A CN202210590392A CN115098766B CN 115098766 B CN115098766 B CN 115098766B CN 202210590392 A CN202210590392 A CN 202210590392A CN 115098766 B CN115098766 B CN 115098766B
- Authority
- CN
- China
- Prior art keywords
- keyword
- user
- information
- data
- target
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/08—Auctions
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a bidding information recommendation method and system for an electronic bidding transaction platform, wherein the method comprises the steps of receiving target advertising information on the transaction platform, carrying out matching search by combining a preset keyword word library based on the target advertising information, and obtaining first keyword data of the target advertising information; generating second keyword data of the target bulletin information based on analysis of the target bulletin information when the matching search result is empty; analyzing the interest degree parameters of each keyword to the user based on the first keyword data or the second keyword data of the target bulletin information, wherein the interest degree parameters of each keyword to the user are obtained based on interest degree labels of the user to the keywords; recommending the target advertisement information to the user based on the interestingness ranking. The invention realizes the rapid and intelligent matching of the two parties of the acquisition and supply, helps the two parties of the bidding to acquire effective information, and ensures efficient and accurate transaction behaviors.
Description
Technical Field
The invention relates to the technical field of bidding, in particular to a bidding information recommendation method and system for an electronic bidding trading platform
Background
The bidding and bidding are used as wide-range and general market trading behaviors, the monopolization and regional blocking of the industry are broken, the electronic bidding and bidding enables the bidding behaviors to be fairer, more fair and public, the gray zone of the industry is reduced, the manpower, material resources and financial resources are saved, and the cost reduction and synergy are realized. Along with the electronization of bidding business and enterprise purchasing business, various electronic bidding transaction platforms emerge, various enterprises independently produce transaction platforms, and form the phenomenon of information overload, so that the phenomenon of more, messy and miscellaneous bidding information on the Internet, namely more bidding bulletin release platforms, messy bidding bulletin formats, messy bidding bulletin contents and distribution, can prevent bidding people from rapidly and effectively acquiring information.
The tenderer depends on past cooperative bidders on one hand, and on active bidding of the bidders on the other hand. For bidders, on one hand, the bidders actively search whether the tenderer concerned by the bidders issues projects or not, and on the other hand, the bidders search related products of the tenderer in various electronic tendering platforms, so that proper bidding opportunities are searched. The bidding parties have the problems of missing bidding information, difficult entry, insufficient bidders, unsuitable bidders and the like.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a bidding information recommendation method and a bidding information recommendation system for an electronic bidding transaction platform, which effectively improve the rapid and intelligent matching of the two parties.
In a first aspect, the present invention provides a bidding information recommendation method for an electronic bidding transaction platform, including:
receiving target bulletin information on a transaction platform, and carrying out matching search by combining a preset keyword word stock based on the target bulletin information to obtain first keyword data of the target bulletin information;
generating second keyword data of the target bulletin information based on analysis of the target bulletin information when the matching search result is empty;
analyzing the interest degree parameters of each keyword to the user based on the first keyword data or the second keyword data of the target bulletin information, wherein the interest degree parameters of each keyword to the user are obtained based on interest degree labels of the user to the keywords;
recommending the target advertisement information to the user based on the interestingness ranking.
In some embodiments, the performing a matching search based on the target advertisement information in combination with a preset keyword lexicon to obtain first keyword data of the target advertisement information includes:
extracting keyword labels based on the historical bulletin information, and establishing a preset keyword lexicon based on the extracted keywords;
data cleaning is carried out on the target bulletin information, and company names, regional names and person names in the target bulletin are removed;
and searching in a preset keyword word stock in a matched mode based on the cleaned target bulletin information, and acquiring first keyword data of the target bulletin information.
In some embodiments, the generating the second keyword data of the target advertisement information based on analyzing the target advertisement information itself includes:
based on the history bulletin information, calculating high-frequency words appearing in the history bulletin through a first preset algorithm, and establishing a white list word stock based on the high-frequency words;
acquiring verbs and nouns in the target bulletin information by using a second preset algorithm based on the cleaned target bulletin information to form a first proper noun set;
filtering and screening verbs and nouns in the first proper noun set by adopting a white list word stock to form a second proper noun set;
and combining the verbs and nouns in the second proper noun set to form second keyword data.
In some embodiments, the analyzing the interest degree parameter of each keyword to the user based on the first keyword data or the second keyword data of the target advertisement information includes:
acquiring a mapping relation between the operation behavior of a user on the bulletin information and the interest degree of the user on the bulletin information, which is established in advance;
extracting first keyword data or second keyword data for the history announcement information;
acquiring interest degree parameters of the user on different keyword data of the historical bulletin information based on the operation behaviors of the user on the bulletin information in the historical record and the mapping relation;
constructing a user keyword interest portraits based on interest degree parameters of the user on different keyword data of the history bulletin information;
and comparing and analyzing the first keyword data or the second keyword data based on the target bulletin information with the user keyword interest portraits, and analyzing the fusion interest degree parameters of each keyword to the user.
In some embodiments, the obtaining the interestingness parameter of the user for different keyword data of the historical bulletin information based on the operation behavior of the user for bulletin information in the history record and the mapping relation includes:
acquiring first interestingness parameters of different keyword data of the historical bulletin information by a user based on the operation behaviors of the user on the bulletin information in the historical record and the mapping relation;
dividing different keywords with the same first interestingness parameter into the same set, marking the first set and marking the first set with the first interestingness parameter;
text semantic recognition and classification are carried out based on keywords in the same first set, and a plurality of first subclasses in the same first set are obtained;
performing similarity analysis based on the first subclasses in all the first sets, and secondarily analyzing and correcting the belonging sets of the different first subclasses of the different sets when the similarity of the different first subclasses distributed in the different sets is smaller than a first preset similarity threshold;
and labeling a second interestingness parameter for the keywords in the set based on the first interestingness parameter labeled by the set based on the corrected set.
In some embodiments, when the similarity of the first subclass and the second subclass distributed in two different sets is smaller than a first preset similarity threshold, performing secondary analysis on the first subclass and the second subclass to correct the belonging sets of the first subclass and the second subclass, including:
based on the first subclasses in all the first sets, for different first subclasses with similarity smaller than a first preset similarity threshold distributed in different sets, acquiring time of operating behaviors of users on the advertising information of the source of the keyword in the different first subclasses, and counting frequency of operating behaviors of the users on the advertising information of the source of the keyword in the different first subclasses in the same time period;
determining a corrected belonging set of different first subclasses distributed in different sets, wherein the similarity is smaller than a first preset similarity threshold value, based on the time and frequency data of the operation behaviors;
and merging different first subclasses with the distribution of which the similarity is smaller than a first preset similarity threshold value in different sets, and classifying the first subclasses into the corrected set.
In some embodiments, the constructing the user keyword interest portraits based on the user's interest level parameters for different keyword data of the historical bulletin information includes:
the method comprises the steps of sorting keyword data according to the time of operation behaviors of a user on bulletin information corresponding to the keyword data;
for the sorted keyword data, taking the interestingness parameter as first annotation data, taking the time data as second annotation data, classifying the keyword data based on the advertisement information type characterized by the keyword data, generating third annotation data, and generating a user keyword interest portrait based on the first annotation data, the second annotation data and the third annotation data, wherein the weight of each keyword in the user keyword interest portrait is determined based on the first annotation data, the second annotation data and the third annotation data, and the weight of each keyword in the user keyword interest portrait is as follows:wherein omega 0 Is a weight basic value, t is the current second labeling data, n is the number of keyword data with the same third labeling data, t s Second labeling data of the s-th keyword data among the keyword data having the same third labeling data, t 0 Is a decay factor of user keyword interests over time.
In some embodiments, the comparing and analyzing the first keyword data or the second keyword data based on the target advertisement information and the interest portraits of the keywords of the user, analyzing the fusion interestingness parameter of each keyword to the user, including:
acquiring the weight of a keyword in a user keyword interest portrait;
searching the times and the positions of occurrence of the keyword data in the user keyword interest portraits based on the first keyword data or the second keyword data;
and calculating a fused interestingness parameter of the keyword data to the user based on the weight of the keywords of the positions appearing in the user keyword interest portraits.
In a second aspect, the present invention provides a bidding information recommendation system of an electronic bidding transaction platform, comprising:
the first keyword acquisition unit is used for receiving target bulletin information on the transaction platform, carrying out matching search by combining a preset keyword word stock based on the target bulletin information, and acquiring first keyword data of the target bulletin information;
the second keyword acquisition unit is used for generating second keyword data of the target bulletin information based on analysis of the target bulletin information when the matching search result is empty;
the user interest degree parameter calculation unit is used for analyzing the interest degree parameter of each keyword to the user based on the first keyword data or the second keyword data of the target bulletin information, wherein the interest degree parameter of each keyword to the user is obtained based on the interest degree label of the user to the keyword;
and the information recommending unit is used for recommending the target bulletin information to the user based on the interestingness sorting.
In a third aspect, the present invention provides a computer-readable storage medium having stored thereon a program which, when executed by a processor, implements the recommendation method according to the first aspect.
The bidding information recommendation method and system for the electronic bidding transaction platform have the following beneficial effects: through the two keyword data acquisition methods, the feature data extraction effectiveness of target bulletin information is improved, meanwhile, target bulletin information and users are analyzed based on keyword data of the target bulletin information and interestingness parameters of each keyword to obtain bidding information recommendation results, in the embodiment of the application, aiming at the problems that bidding information is more, messy and mixed on the Internet, namely, bidding bulletin release platforms are more, bidding bulletin formats are messy, bidding bulletin contents and distribution are messy, rapid and effective information acquisition of bidding persons is hindered, analysis is carried out on user behaviors, user preferences and the like, rapid and intelligent matching of two parties is realized, the bidding parties are helped to acquire effective information, and efficient and accurate transaction behaviors are ensured.
Drawings
FIG. 1 is a flow chart of a method for recommending bidding information for an electronic bidding transaction platform in an embodiment of the present application;
FIG. 2 is a flowchart illustrating a process of analyzing a fused interestingness parameter of each keyword of target advertisement information to a user according to an embodiment of the present application;
FIG. 3 is a flowchart of obtaining interest level parameters of users for different keyword data of historical bulletin information according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of a bidding information recommendation system of the electronic bidding transaction platform in the embodiment of the present application.
Detailed Description
The present invention will be further described in detail with reference to the accompanying drawings, for the purpose of making the objects, technical solutions and advantages of the present invention more apparent, and the described embodiments should not be construed as limiting the present invention, and all other embodiments obtained by those skilled in the art without making any inventive effort are within the scope of the present invention.
The embodiment of the application provides a bidding information recommendation method of an electronic bidding transaction platform, which comprises the following steps:
and step 4, recommending the target bulletin information to the user based on the interestingness sorting.
According to the embodiment of the application, the effectiveness of extracting the characteristic data of the target bulletin information is improved through the two keyword data acquisition methods, meanwhile, the target bulletin information and the user are analyzed based on the keyword data of the target bulletin information and the interestingness parameter of each keyword, and the bidding information recommendation result is obtained.
Further, in the step 1, performing a matching search based on the target advertisement information in combination with a preset keyword word stock to obtain first keyword data of the target advertisement information, including:
extracting keyword labels based on the historical bulletin information, and establishing a preset keyword lexicon based on the extracted keywords;
data cleaning is carried out on the target bulletin information, and company names, regional names and person names in the target bulletin are removed;
and searching in a preset keyword word stock in a matched mode based on the cleaned target bulletin information, and acquiring first keyword data of the target bulletin information.
In the embodiment of the application, a keyword word stock is constructed based on historical data, and word in target bulletin information is matched and searched in a preset keyword word stock, so that word segmentation and keyword screening of the target bulletin information are realized.
Further, in the step 2, generating the second keyword data of the target advertisement information based on the analysis of the target advertisement information itself includes:
based on the history bulletin information, calculating high-frequency words appearing in the history bulletin through a first preset algorithm, and establishing a white list word stock based on the high-frequency words, wherein the first preset algorithm is realized based on an ITF algorithm;
acquiring verbs and nouns in the target bulletin information by using a second preset algorithm based on the cleaned target bulletin information to form a first proper noun set, wherein the second preset algorithm is realized based on a Hadamard word segmentation device;
filtering and screening verbs and nouns in the first proper noun set by adopting a white list word stock to form a second proper noun set;
and combining the verbs and nouns in the second proper noun set to form second keyword data.
In the embodiment of the present invention, another method for extracting second keyword data of target advertisement information is provided for the case that keyword matching cannot be performed on target advertisement information through a preset keyword lexicon established in advance to obtain keyword data in the target advertisement information, and under the condition that a white list lexicon is established through high-frequency words in history advertisement information, proper nouns in the target advertisement information are extracted and filtered in real time and words in the white list lexicon are reserved to form a second proper noun set and second keyword data, so that success rate and comprehensiveness of extracting the keyword data of the target advertisement information are improved.
Further, in the step 3, based on the first keyword data or the second keyword data of the target advertisement information, the step of analyzing the interest degree parameter of each keyword to the user includes:
and 35, comparing and analyzing the first keyword data or the second keyword data of the target bulletin information with the interest portraits of the keywords of the user, and analyzing the fusion interestingness parameters of each keyword to the user.
In the embodiment of the application, through the historical bulletin information and the operation behaviors of the user on the bulletin information in the historical record, the keyword interest portraits of the user are generated based on the different interest degree data of the user, and then whether the target bulletin information is recommended to the user is analyzed based on the keyword interest portraits of the user.
Further, in the step 33, based on the operation behavior of the user on the advertisement information in the history record and the mapping relationship, the obtaining the interestingness parameter of the user on the different keyword data of the history advertisement information includes:
It will be appreciated that when the user operates the advertisement information in the history, the degree of interest of the same type of advertisement information may be different in different periods, for example, the most interest in engineering construction and building material decoration is generated in a certain period, the bidding operation is frequently generated, the degree of interest in electric power electronic class is the largest in a certain period, the bidding operation is frequently generated, the degree of interest in engineering construction and building material decoration is not as great as that of electric power electronic class, and only click browsing operation is performed.
For this case, in the present application, different keywords with the same first interestingness parameter are divided into the same set, and keywords in the same first set are further classified into a plurality of first subclasses, whether the type corresponding to each first subclass and the similar type thereof are simultaneously present in a plurality of different first sets is measured, if so, the types corresponding to the first subclasses simultaneously present in the plurality of different first sets and the similar types thereof are combined and reclassified into the modified belonging set. It can be understood that in the embodiment of the application, the interest degree parameters of the user for obtaining the different keyword data of the historical bulletin information are updated in real time based on the operation behaviors recently happened by the user, so that the dynamic interest degree parameter classification of the user keywords is realized. Furthermore, the interest portraits of the user keywords are constructed based on the interest degree parameters of the user on different keyword data of the historical bulletin information, and the constructed interest portraits of the user keywords are dynamically changed in real time, so that the real-time performance and the effectiveness of the construction of the interest portraits of the user keywords are realized.
Further, in the step 334, when the similarity between the first sub-class and the second sub-class distributed in the two different sets is smaller than the first preset similarity threshold, performing secondary analysis on the first sub-class and the second sub-class to correct the belonging sets of the first sub-class and the second sub-class, including:
step 3341, based on the first subclasses in all the first sets, for different first subclasses with similarity smaller than a first preset similarity threshold distributed in different sets, obtaining the time of the user to operate the advertising information of the source of the keyword in the different first subclasses, and counting the frequency of the user to operate the advertising information of the source of the keyword in the different first subclasses in the same time period;
step 3342, determining, based on the time and frequency data of the occurrence of the operation behavior, a modified belonging set of different first subclasses distributed in different sets, wherein the similarity is smaller than a first preset similarity threshold;
step 3343, merging the different first subclasses distributed in different sets with similarity smaller than the first preset similarity threshold, and classifying the first subclasses into the corrected set.
It may be understood that, based on the time and frequency data of the occurrence of the operation behavior, the corrected set of different first sub-classes with the similarity smaller than the first preset similarity threshold may be a corrected set of the first sub-class with the greater frequency data in the same time period, which is closer to the current time, as the corrected set of the first sub-class, and the closer to the current time, the closer to the recent interest degree of the user in the keyword, the greater frequency data in the same time period indicates that the operation behavior of the user indicates the interest degree of the user, and it may be understood that if the frequency data in the same time period is very small, it indicates that the corresponding operation behavior of the user is accidental such as misoperation, and the interest degree of the user cannot be indicated.
Further, in the step 34, the step of constructing the user keyword interest portraits based on the interest level parameters of the user on the different keyword data of the history announcement information includes:
step 341, sorting the keyword data according to the time when the user performs operation on the advertisement information corresponding to the keyword data;
step 342, for the sorted keyword data, classifying the keyword data based on the interestingness parameter as first labeling data, time data as second labeling data, and advertisement information type represented by the keyword data, generating third labeling data, and generating a user keyword interest portrait based on the first labeling data, the second labeling data and the third labeling data, wherein the weight of each keyword in the user keyword interest portrait is determined based on the first labeling data, the second labeling data and the third labeling data, and the weight of each keyword in the user keyword interest portrait is as follows:wherein omega 0 Is a weight basic value, t is the current second labeling data, n is the number of keyword data with the same third labeling data, t s Second labeling data of the s-th keyword data among the keyword data having the same third labeling data, t 0 Is a decay factor of user keyword interests over time.
Further, in the step 35, based on the first keyword data or the second keyword data of the target advertisement information and the interest portraits of the keywords of the user, a fused interestingness parameter of each keyword to the user is analyzed, including:
step 351, obtaining keywords and weights of the keywords in the user keyword interest portraits;
step 352, searching the user keyword interest portraits for the number of times and the appearance position of the keyword data in the user keyword interest portraits based on the first keyword data or the second keyword data;
step 353, calculating a fused interestingness parameter of the keyword data to the user based on the weights of keywords at locations where the user keyword interest portraits appear.
Based on the bidding information recommendation method of the electronic bidding transaction platform, the embodiment of the application also provides a bidding information recommendation system of the electronic bidding transaction platform, which comprises the following steps:
the first keyword acquisition unit is used for receiving target bulletin information on the transaction platform, carrying out matching search by combining a preset keyword word stock based on the target bulletin information, and acquiring first keyword data of the target bulletin information;
the second keyword acquisition unit is used for generating second keyword data of the target bulletin information based on analysis of the target bulletin information when the matching search result is empty;
the user interest degree parameter calculation unit is used for analyzing the interest degree parameter of each keyword to the user based on the first keyword data or the second keyword data of the target bulletin information, wherein the interest degree parameter of each keyword to the user is obtained based on the interest degree label of the user to the keyword;
and the information recommending unit is used for recommending the target bulletin information to the user based on the interestingness sorting.
Specific limitations regarding the electronic bidding transaction platform bidding information recommendation system may be found in the above limitations regarding the electronic bidding transaction platform bidding information recommendation method, and will not be described herein. Each unit in the electronic bidding transaction platform bidding information recommendation system described above may be implemented in whole or in part by software, hardware, and combinations thereof. The units can be embedded in hardware or independent of a processor in the computer equipment, and can also be stored in a memory in the computer equipment in a software mode, so that the processor can call and execute the operations corresponding to the units.
Based on the electronic bidding transaction platform bidding information recommendation method, the embodiment of the application also provides a computer readable storage medium, wherein a program is stored on the computer readable storage medium, and the program is executed by a processor to realize the electronic bidding transaction platform bidding information recommendation method.
Specifically, the computer-readable storage medium includes: persistent and non-persistent, removable and non-removable media are tangible devices that may retain and store instructions for use by an instruction execution device. The computer-readable storage medium includes: electronic storage, magnetic storage, optical storage, electromagnetic storage, semiconductor storage, and any suitable combination of the foregoing.
The present invention is not limited to the above-described specific embodiments, and various modifications may be made by those skilled in the art without inventive effort from the above-described concepts, and are within the scope of the present invention.
Claims (8)
1. A bidding information recommendation method for an electronic bidding transaction platform, comprising:
receiving target bulletin information on a transaction platform, and carrying out matching search by combining a preset keyword word stock based on the target bulletin information to obtain first keyword data of the target bulletin information;
generating second keyword data of the target bulletin information based on analysis of the target bulletin information when the matching search result is empty;
analyzing the interest degree parameters of each keyword to the user based on the first keyword data or the second keyword data of the target bulletin information, wherein the interest degree parameters of each keyword to the user are obtained based on interest degree labels of the user to the keywords;
recommending the target bulletin information to the user based on the interestingness ranking;
the first keyword data or the second keyword data based on the target bulletin information, analyze the interestingness parameter of each keyword to the user, including:
acquiring a mapping relation between the operation behavior of a user on the bulletin information and the interest degree of the user on the bulletin information, which is established in advance;
extracting first keyword data or second keyword data for the history announcement information;
acquiring interest degree parameters of the user on different keyword data of the historical bulletin information based on the operation behaviors of the user on the bulletin information in the historical record and the mapping relation;
constructing a user keyword interest portraits based on interest degree parameters of the user on different keyword data of the history bulletin information;
based on the first keyword data or the second keyword data of the target bulletin information and the interest portraits of the keywords of the user, carrying out comparative analysis, and analyzing the fusion interest degree parameters of each keyword to the user;
the step of obtaining the interestingness parameter of the user for different keyword data of the historical bulletin information based on the operation behavior of the user for the bulletin information in the historical record and the mapping relation comprises the following steps:
acquiring first interestingness parameters of different keyword data of the historical bulletin information by a user based on the operation behaviors of the user on the bulletin information in the historical record and the mapping relation;
dividing different keywords with the same first interestingness parameter into the same set, marking the first set and marking the first set with the first interestingness parameter;
text semantic recognition and classification are carried out based on keywords in the same first set, and a plurality of first subclasses in the same first set are obtained;
performing similarity analysis based on the first subclasses in all the first sets, and secondarily analyzing and correcting the belonging sets of the different first subclasses of the different sets when the similarity of the different first subclasses distributed in the different sets is smaller than a first preset similarity threshold;
and labeling a second interestingness parameter for the keywords in the set based on the first interestingness parameter labeled by the set based on the corrected set.
2. The method for recommending bidding information for an electronic bidding transaction platform according to claim 1, wherein the performing a matching search based on the target advertisement information in combination with a preset keyword word library to obtain first keyword data of the target advertisement information comprises:
extracting keyword labels based on the historical bulletin information, and establishing a preset keyword lexicon based on the extracted keywords;
data cleaning is carried out on the target bulletin information, and company names, regional names and person names in the target bulletin are removed;
and searching in a preset keyword word stock in a matched mode based on the cleaned target bulletin information, and acquiring first keyword data of the target bulletin information.
3. The method for recommending bid information for an electronic bidding transaction platform according to claim 1, wherein the generating the second keyword data of the target advertisement information based on the analysis of the target advertisement information itself comprises:
based on the history bulletin information, calculating high-frequency words appearing in the history bulletin through a first preset algorithm, and establishing a white list word stock based on the high-frequency words;
acquiring verbs and nouns in the target bulletin information by using a second preset algorithm based on the cleaned target bulletin information to form a first proper noun set;
filtering and screening verbs and nouns in the first proper noun set by adopting a white list word stock to form a second proper noun set;
and combining the verbs and nouns in the second proper noun set to form second keyword data.
4. The method for recommending bidding information for an electronic bidding transaction platform according to claim 1, wherein when the similarity of the first subclass and the second subclass distributed in two different sets is smaller than a first preset similarity threshold, performing secondary analysis on the first subclass and the second subclass to correct the belonging sets of the first subclass and the second subclass, comprising:
based on the first subclasses in all the first sets, for different first subclasses with similarity smaller than a first preset similarity threshold distributed in different sets, acquiring time of operating behaviors of users on the advertising information of the source of the keyword in the different first subclasses, and counting frequency of operating behaviors of the users on the advertising information of the source of the keyword in the different first subclasses in the same time period;
determining a corrected belonging set of different first subclasses distributed in different sets, wherein the similarity is smaller than a first preset similarity threshold value, based on the time and frequency data of the operation behaviors;
and merging different first subclasses with the distribution of which the similarity is smaller than a first preset similarity threshold value in different sets, and classifying the first subclasses into the corrected set.
5. The method for recommending bidding information for an electronic bidding transaction platform according to claim 1, wherein the constructing the user keyword interest portraits based on the user interest level parameters for different keyword data of the historical advertising information comprises:
the method comprises the steps of sorting keyword data according to the time of operation behaviors of a user on bulletin information corresponding to the keyword data;
for the sorted keyword data, taking the interestingness parameter as first annotation data, taking the time data as second annotation data, classifying the keyword data based on the advertisement information type characterized by the keyword data, generating third annotation data, and generating a user keyword interest portrait based on the first annotation data, the second annotation data and the third annotation data, wherein the weight of each keyword in the user keyword interest portrait is determined based on the first annotation data, the second annotation data and the third annotation data, and the weight of each keyword in the user keyword interest portrait is as follows:wherein, the method comprises the steps of, wherein,is the weight base value, ++>Is the current second annotation data, +.>Is the number of keyword data having the same third label data,/-, and>is the first +.in the keyword data with the same third labeling data>Second labeling data of the individual keyword data, < ->Is a decay factor of user keyword interests over time.
6. The method for recommending bidding information for an electronic bidding transaction platform according to claim 1, wherein the comparing and analyzing the first keyword data or the second keyword data based on the target advertisement information with the user keyword interest portraits, analyzing the fused interestingness parameter of each keyword for the user, comprises:
acquiring the weight of a keyword in a user keyword interest portrait;
searching the times and the positions of occurrence of the keyword data in the user keyword interest portraits based on the first keyword data or the second keyword data;
and calculating a fused interestingness parameter of the keyword data to the user based on the weight of the keywords of the positions appearing in the user keyword interest portraits.
7. An electronic bidding transaction platform bidding information recommendation system, comprising:
the first keyword acquisition unit is used for receiving target bulletin information on the transaction platform, carrying out matching search by combining a preset keyword word stock based on the target bulletin information, and acquiring first keyword data of the target bulletin information;
the second keyword acquisition unit is used for generating second keyword data of the target bulletin information based on analysis of the target bulletin information when the matching search result is empty;
the user interest degree parameter calculation unit is used for analyzing the interest degree parameter of each keyword to the user based on the first keyword data or the second keyword data of the target bulletin information, wherein the interest degree parameter of each keyword to the user is obtained based on the interest degree label of the user to the keyword;
an information recommending unit for recommending the target advertisement information to the user based on the interestingness ranking;
the user interest degree parameter calculating unit analyzes the interest degree parameter of each keyword to the user based on the first keyword data or the second keyword data of the target bulletin information, and comprises:
acquiring a mapping relation between the operation behavior of a user on the bulletin information and the interest degree of the user on the bulletin information, which is established in advance;
extracting first keyword data or second keyword data for the history announcement information;
acquiring interest degree parameters of the user on different keyword data of the historical bulletin information based on the operation behaviors of the user on the bulletin information in the historical record and the mapping relation;
constructing a user keyword interest portraits based on interest degree parameters of the user on different keyword data of the history bulletin information;
based on the first keyword data or the second keyword data of the target bulletin information and the interest portraits of the keywords of the user, carrying out comparative analysis, and analyzing the fusion interest degree parameters of each keyword to the user;
the step of obtaining the interestingness parameter of the user for different keyword data of the historical bulletin information based on the operation behavior of the user for the bulletin information in the historical record and the mapping relation comprises the following steps:
acquiring first interestingness parameters of different keyword data of the historical bulletin information by a user based on the operation behaviors of the user on the bulletin information in the historical record and the mapping relation;
dividing different keywords with the same first interestingness parameter into the same set, marking the first set and marking the first set with the first interestingness parameter;
text semantic recognition and classification are carried out based on keywords in the same first set, and a plurality of first subclasses in the same first set are obtained;
performing similarity analysis based on the first subclasses in all the first sets, and secondarily analyzing and correcting the belonging sets of the different first subclasses of the different sets when the similarity of the different first subclasses distributed in the different sets is smaller than a first preset similarity threshold;
and labeling a second interestingness parameter for the keywords in the set based on the first interestingness parameter labeled by the set based on the corrected set.
8. A computer readable storage medium having stored thereon a program which, when executed by a processor, implements the recommendation method according to any of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210590392.6A CN115098766B (en) | 2022-05-26 | 2022-05-26 | Bidding information recommendation method and system for electronic bidding transaction platform |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210590392.6A CN115098766B (en) | 2022-05-26 | 2022-05-26 | Bidding information recommendation method and system for electronic bidding transaction platform |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115098766A CN115098766A (en) | 2022-09-23 |
CN115098766B true CN115098766B (en) | 2023-05-30 |
Family
ID=83288274
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210590392.6A Active CN115098766B (en) | 2022-05-26 | 2022-05-26 | Bidding information recommendation method and system for electronic bidding transaction platform |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115098766B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115905489B (en) * | 2022-11-21 | 2023-11-17 | 广西建设职业技术学院 | Method for providing bidding information search service |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109165348A (en) * | 2018-08-21 | 2019-01-08 | 麦格创科技(深圳)有限公司 | A kind of bidding information recommendation method, system and server |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109446407A (en) * | 2018-09-18 | 2019-03-08 | 深圳壹账通智能科技有限公司 | Correlation recommendation method, apparatus, computer equipment and storage medium |
CN110148043A (en) * | 2019-03-01 | 2019-08-20 | 安徽省优质采科技发展有限责任公司 | The bid and purchase information recommendation system and recommended method of knowledge based map |
CN110059246A (en) * | 2019-03-15 | 2019-07-26 | 安徽省优质采科技发展有限责任公司 | Intelligent match system |
GB2590206A (en) * | 2019-04-30 | 2021-06-23 | Beijing Bytedance Network Tech Co Ltd | Object recommendation method and apparatus, storage medium, and terminal device |
-
2022
- 2022-05-26 CN CN202210590392.6A patent/CN115098766B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109165348A (en) * | 2018-08-21 | 2019-01-08 | 麦格创科技(深圳)有限公司 | A kind of bidding information recommendation method, system and server |
Also Published As
Publication number | Publication date |
---|---|
CN115098766A (en) | 2022-09-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109189942B (en) | Construction method and device of patent data knowledge graph | |
CN112184525B (en) | System and method for realizing intelligent matching recommendation through natural semantic analysis | |
CN110263248B (en) | Information pushing method, device, storage medium and server | |
CN112035742B (en) | User portrait generation method, device, equipment and storage medium | |
CN108363821A (en) | A kind of information-pushing method, device, terminal device and storage medium | |
CN111552870A (en) | Object recommendation method, electronic device and storage medium | |
CN109819015B (en) | Information pushing method, device and equipment based on user portrait and storage medium | |
CN111538901B (en) | Article recommendation method and device, server and storage medium | |
CN108984554B (en) | Method and device for determining keywords | |
CN111553137B (en) | Report generation method and device, storage medium and computer equipment | |
JP2015525418A (en) | Search method and apparatus | |
CN113901308A (en) | Knowledge graph-based enterprise recommendation method and recommendation device and electronic equipment | |
CN111782793A (en) | Intelligent customer service processing method, system and equipment | |
CN111027838A (en) | Crowdsourcing task pushing method, device, equipment and storage medium thereof | |
CN115098766B (en) | Bidding information recommendation method and system for electronic bidding transaction platform | |
CN115391669A (en) | Intelligent recommendation method and device and electronic equipment | |
CN112989824A (en) | Information pushing method and device, electronic equipment and storage medium | |
CN108875014B (en) | Precise project recommendation method based on big data and artificial intelligence and robot system | |
CN112487808A (en) | Big data based news message pushing method, device, equipment and storage medium | |
KR102404247B1 (en) | Customer management system | |
CN114282119B (en) | Scientific and technological information resource retrieval method and system based on heterogeneous information network | |
CN114996579A (en) | Information pushing method and device, electronic equipment and computer readable medium | |
CN114201680A (en) | Method for recommending marketing product content to user | |
CN115344774A (en) | User account screening method and device and server | |
CN112818215A (en) | Product data processing method, device, equipment and storage medium |
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 | ||
CB02 | Change of applicant information |
Address after: Room 201, Building A11, Financial Port Center, No. 4872 Huizhou Avenue, Baohe District, Hefei City, Anhui Province, 230092 Applicant after: Anhui High Quality Mining Technology Development Co.,Ltd. Address before: Room 401, Yunhui Pavilion Commercial Complex, Intersection of Baohe Avenue and Wangnan Road, Baohe District, Hefei City, Anhui Province 230041 Applicant before: Anhui High Quality Mining Technology Development Co.,Ltd. |
|
CB02 | Change of applicant information | ||
GR01 | Patent grant | ||
GR01 | Patent grant |