Recommendation method, recommendation system, computer equipment and medium for construction engineering information
By cleaning and building correlation maps of construction engineering information, and building recommendation models with user data, the problem of difficulty in identifying business opportunities is solved, and more accurate and efficient business opportunity recommendations are achieved.
Patent Information
- Application Number
- CN202410648940.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-23
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-05-23
AI Technical Summary
When building materials companies obtain construction project information, it is difficult for them to identify potential business opportunities and project needs, resulting in reduced operational efficiency.
By obtaining the construction project information to be analyzed, cleaning the data, building a project information correlation map, and combining the user's subscription data and behavioral data, a project information recommendation model is built to generate personalized recommendation results in real time.
It improves the accuracy and liquidity of construction project information recommendation, helps building materials companies identify potential business opportunities and project needs, and improves the company's operational efficiency.
Smart Images

Figure CN118469228B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of industrial big data, and in particular to a method, a system, a computer device and a medium for recommending construction engineering information. Background Art
[0002] With the rapid development of the construction industry in recent years, competition in the construction engineering market has become increasingly fierce. Due to the wide variety and complexity of construction engineering information, some building materials companies, after obtaining a large amount of engineering information through various media channels, field visits and face-to-face interviews, will have problems such as not knowing which data to look at, not finding key business opportunities, not being able to judge whether the company is worth following up, and not having a deep understanding of the information. This reduces the liquidity of construction engineering information, making it impossible for building materials companies to effectively identify potential business opportunities and project needs, and reducing the operating efficiency of building materials companies. Summary of the invention
[0003] The purpose of the present invention is to solve at least one of the technical problems existing in the prior art, and to provide a method, system, computer equipment and medium for recommending construction engineering information, which can provide more reliable data support for corresponding building materials companies to identify potential business opportunities and project needs, thereby helping to improve the operational efficiency of the companies.
[0004] In a first aspect, an embodiment of the present invention provides a method for recommending construction project information, comprising:
[0005] Obtain the construction project information to be analyzed;
[0006] Performing data cleaning on the construction project information to be analyzed to obtain target construction project information, wherein the target construction project information includes a plurality of target data entities;
[0007] Constructing an engineering information association map based on the multiple target data entities, wherein the engineering information association map is used to reflect the relationship between each of the target data entities;
[0008] Obtain user subscription data and behavior data for construction projects;
[0009] Building a project information recommendation model according to the subscription data, the behavior data and the project information association graph;
[0010] In response to a behavior instruction generated by a user visiting a target project page, a recommendation result is generated according to the project information recommendation model.
[0011] The method for recommending construction project information provided by the embodiment of the present invention has at least the following beneficial effects: by acquiring the construction project information to be analyzed, performing data cleaning on the construction project information to be analyzed, and obtaining the target construction project information, the data quality and accuracy of the target construction project information can be guaranteed; the target construction project information includes multiple target data entities; by constructing a project information association map based on multiple target data entities, the relationship between each target data entity can be reflected, which is convenient for subsequent in-depth data mining and analysis; by acquiring the user's subscription data and behavior data for construction projects, the user's preferences can be better reflected; a project information recommendation model is constructed based on subscription data, behavior data and project information association maps, which can improve the accuracy of the project information recommendation model; in response to the behavior instructions generated by the user visiting the target project page, a personalized recommendation result is generated in real time using the project information recommendation model, which can provide users with more accurate business opportunity recommendations; it can be understood that the present invention can enhance the fluidity and depth of data by constructing a project information association map using applied statistics and data analysis technology; at the same time, combined with the user's preferences, it can provide more reliable data support for the corresponding building materials companies to identify potential business opportunities and project requirements, which is conducive to improving the operational efficiency of the company.
[0012] In the above-mentioned construction project information recommendation method, the data cleaning of the construction project information to be analyzed to obtain the target construction project information includes:
[0013] Traversing the construction project information to be analyzed, identifying and deleting duplicate information, and obtaining first preprocessing information;
[0014] Performing unified format processing on the first preprocessing information according to a preset format rule to obtain second preprocessing information;
[0015] Performing missing value processing and outlier processing on the second preprocessed information to obtain third preprocessed information;
[0016] Performing desensitization processing on the third preprocessing information to obtain fourth preprocessing information;
[0017] The fourth preprocessing information is subjected to data verification according to preset verification rules to obtain target construction project information.
[0018] In the above-mentioned construction project information recommendation method, the step of constructing a project information association graph based on the multiple target data entities includes:
[0019] Constructing entity attribute information according to each of the target data entities, and generating a corresponding entity information table, wherein the entity information table corresponds to the target data entity;
[0020] Determine the relationship type between each of the target data entities, and construct an engineering information association map based on the relationship type.
[0021] In the above-mentioned construction project information recommendation method, the step of constructing a project information recommendation model based on the subscription data, the behavior data and the project information association graph includes:
[0022] Integrate the subscription data, the behavior data and the engineering information association graph to generate a data source;
[0023] Performing feature extraction processing on the data source to convert it into a feature data set, wherein the feature data set is key feature data that affects construction engineering information;
[0024] Determine the data type of the feature data set, and determine the corresponding target recommendation algorithm according to the data type;
[0025] Using the target recommendation algorithm and the feature data set to train an engineering information recommendation model;
[0026] Determining performance evaluation indicators of the engineering information recommendation model;
[0027] Parameters of the engineering information recommendation model are optimized according to the performance evaluation index to obtain a trained engineering information recommendation model.
[0028] In the above-mentioned construction project information recommendation method, the step of determining the data type of the feature data set and determining the corresponding target recommendation algorithm according to the data type includes:
[0029] When it is determined that the feature data set is a first data type, determining that the target recommendation algorithm is a similarity algorithm, wherein the first data type is used to reflect that the feature of the feature data set is obvious;
[0030] When it is determined that the feature data set is a second data type, the target recommendation algorithm is determined to be a deep learning recommendation algorithm, wherein the second data type is used to reflect that the data volume of the feature data set is large.
[0031] In the above-mentioned construction project information recommendation method, the generating of recommendation results according to the project information recommendation model in response to the behavior instruction generated by the user visiting the target project page includes:
[0032] In response to a behavior instruction generated by a user accessing a target project page, obtaining corresponding real-time project information;
[0033] Performing feature extraction processing on the real-time engineering information to obtain real-time feature data;
[0034] The real-time feature data is input into the engineering information recommendation model to obtain a recommendation result, wherein the recommendation result includes at least one of project business opportunity information, procurement business opportunity information, company business opportunity information, and a relationship map.
[0035] The recommended method for the above construction project information also includes:
[0036] Displaying the recommendation results in a user interaction interface;
[0037] Obtaining user feedback on the recommendation result;
[0038] The engineering information recommendation model is optimized according to the feedback information.
[0039] In a second aspect, an embodiment of the present invention provides a construction engineering information recommendation system, including:
[0040] The first acquisition module is used to acquire the construction project information to be analyzed;
[0041] A data cleaning module, used for performing data cleaning on the construction project information to be analyzed to obtain target construction project information, wherein the target construction project information includes a plurality of target data entities;
[0042] An association map construction module, used to construct an engineering information association map based on the multiple target data entities, wherein the engineering information association map is used to reflect the relationship between each of the target data entities;
[0043] The second acquisition module is used to acquire the user's subscription data and behavior data for the construction project;
[0044] A recommendation model building module, used to build a project information recommendation model based on the subscription data, the behavior data and the project information association map;
[0045] The recommendation module is used to generate a recommendation result according to the project information recommendation model in response to the behavior instruction generated by the user visiting the target project page.
[0046] The construction project information recommendation system provided by the embodiment of the present invention has at least the following beneficial effects: by acquiring the construction project information to be analyzed, performing data cleaning on the construction project information to be analyzed, and obtaining the target construction project information, the data quality and accuracy of the target construction project information can be guaranteed; the target construction project information includes multiple target data entities; by constructing a project information association map based on multiple target data entities, the relationship between each target data entity can be reflected, which is convenient for subsequent in-depth data mining and analysis; by acquiring the user's subscription data and behavior data for construction projects, the user's preferences can be better reflected; a project information recommendation model is constructed based on subscription data, behavior data and project information association maps, which can improve the accuracy of the project information recommendation model; in response to the behavior instructions generated by the user visiting the target project page, a personalized recommendation result is generated in real time using the project information recommendation model, which can provide users with more accurate business opportunity recommendations; it can be understood that the present invention can enhance the fluidity and depth of data by constructing a project information association map using applied statistics and data analysis technology; at the same time, combined with the user's preferences, it can provide more reliable data support for the corresponding building materials companies to identify potential business opportunities and project requirements, which is conducive to improving the operational efficiency of the company.
[0047] In a third aspect, an embodiment of the present invention provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the recommendation method described in the embodiment of the first aspect is implemented.
[0048] The computer device provided according to the embodiment of the present invention has at least the following beneficial effects: by acquiring the construction project information to be analyzed, performing data cleaning on the construction project information to be analyzed, and obtaining the target construction project information, the data quality and accuracy of the target construction project information can be guaranteed; the target construction project information includes multiple target data entities; by constructing a project information association map based on multiple target data entities, the relationship between each target data entity can be reflected, which is convenient for subsequent in-depth data mining and analysis; by acquiring the user's subscription data and behavior data for the construction project, the user's preference can be better reflected; a project information recommendation model is constructed according to the subscription data, behavior data and project information association map, which can improve the accuracy of the project information recommendation model; in response to the behavior instruction generated by the user visiting the target project page, the project information recommendation model is used to generate personalized recommendation results in real time, which can provide users with more accurate business opportunity recommendations; it can be understood that the present invention can enhance the fluidity and depth of data by constructing a project information association map using applied statistics and data analysis technology; at the same time, combined with the user's preferences, it can provide more reliable data support for the corresponding building materials enterprises to identify potential business opportunities and project needs, which is conducive to improving the operational efficiency of the enterprise.
[0049] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the recommendation method described in the embodiment of the first aspect above.
[0050] The computer-readable storage medium provided according to the embodiment of the present invention has at least the following beneficial effects: by acquiring the construction project information to be analyzed, performing data cleaning on the construction project information to be analyzed, and obtaining the target construction project information, the data quality and accuracy of the target construction project information can be guaranteed; the target construction project information includes multiple target data entities; by constructing a project information association map based on multiple target data entities, the relationship between each target data entity can be reflected, which is convenient for subsequent in-depth data mining and analysis; by acquiring the user's subscription data and behavior data for the construction project, the user's preference can be better reflected; a project information recommendation model is constructed according to the subscription data, behavior data and project information association map, which can improve the accuracy of the project information recommendation model; in response to the behavior instruction generated by the user visiting the target project page, the project information recommendation model is used to generate personalized recommendation results in real time, which can provide users with more accurate business opportunity recommendations; it can be understood that the present invention can enhance the fluidity and depth of data by constructing a project information association map using applied statistics and data analysis technology; at the same time, combined with the user's preferences, it can provide more reliable data support for the corresponding building materials enterprises to identify potential business opportunities and project needs, which is conducive to improving the operational efficiency of the enterprise.
[0051] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0052] The present invention is further described below in conjunction with the accompanying drawings and embodiments;
[0053] Figure 1 It is a flowchart of a method for recommending construction project information provided in the first embodiment of the present invention.
[0054] Figure 2 is a flow chart of a method for recommending construction engineering information provided by Embodiment 2 of the present invention;
[0055] Figure 3 is a schematic diagram of the structure of a recommendation system for construction engineering information provided by Embodiment 3 of the present invention;
[0056] Figure 4It is a structural diagram of a computer device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION
[0057] This section will describe in detail the specific embodiments of the present invention. The preferred embodiments of the present invention are shown in the accompanying drawings. The purpose of the accompanying drawings is to supplement the description of the text part of the specification with graphics, so that people can intuitively and vividly understand each technical feature and the overall technical solution of the present invention, but it cannot be understood as a limitation on the scope of protection of the present invention.
[0058] It should be understood that in the description of the embodiments of the present invention, if there is a description of "first", "second", etc., it is only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features. "At least one" means one or more, "more than" means more than two, greater than, less than, exceeding, etc. are understood to not include the number itself, above, below, within, etc. are understood to include the number itself, and "several" means one or more, unless otherwise clearly and specifically limited. "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. It can be understood that A and / or B can represent the situation where A exists alone, A and B exist at the same time, or B exists alone. Among them, A and B can be singular or plural.
[0059] In addition, unless otherwise clearly specified and limited, the term "connected / connected" should be understood in a broad sense, for example, it can be a fixed connection or a movable connection, a detachable connection or a non-detachable connection, or an integral connection; it can be a mechanical connection, an electrical connection, or can communicate with each other; it can be directly connected or indirectly connected through an intermediate medium. It should be noted that although the logical order is shown in the flowchart, in some cases, the steps shown or described can be performed in a different order from that in the flowchart. It should be noted that the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.
[0060] The embodiments of the present invention are further described below in conjunction with the accompanying drawings.
[0061] like Figure 1 and Figure 2 As shown, an embodiment of the first aspect of the present invention provides a method for recommending construction engineering information, the recommendation method includes but is not limited to steps S110 to S160:
[0062] Step S110: Obtaining the construction project information to be analyzed;
[0063] It should be noted that the construction project information to be analyzed can be obtained from multiple data sources such as project information developers, material suppliers, databases, etc. The construction project information to be analyzed includes but is not limited to project information, company information, contact information, etc.
[0064] Step S120: performing data cleaning on the construction project information to be analyzed to obtain target construction project information, wherein the target construction project information includes a plurality of target data entities;
[0065] In one embodiment, multiple target data entities may be defined according to business requirements, such as engineering projects, companies, contacts, etc.
[0066] Step S130: constructing an engineering information association map based on multiple target data entities, wherein the engineering information association map is used to reflect the relationship between each target data entity;
[0067] By constructing a project information association map, multiple target data entities can be associated, which facilitates in-depth data mining and analysis and discovers hidden patterns and associations between data. In one embodiment, the project information association map includes detailed information on the project, relationships between companies, contact information, etc.
[0068] Step S140: Obtaining the user's subscription data and behavior data for the construction project;
[0069] Step S150: constructing a project information recommendation model based on subscription data, behavior data and project information association graph;
[0070] When building a project information recommendation model, user subscription data and behavior data are two key data types that are essential for understanding user preferences and providing personalized recommendations. Subscription data refers to the content that users actively choose to subscribe to or follow for construction projects. For example, subscription data includes information about users' attention to certain projects or companies. User behavior data refers to the user's operation records in the corresponding construction project query service or application. For example, behavior data includes user clicks, browsing, searches, and other operation records on the corresponding construction project service system.
[0071] Step S160: In response to the behavior instruction generated by the user visiting the target project page, a recommendation result is generated according to the project information recommendation model.
[0072] When a user visits the target project page, a behavior instruction can be triggered, so that the user's behavior can be tracked in real time, and the recommendation results corresponding to the behavior instruction can be generated according to the project information recommendation model, so as to provide users with accurate business opportunity recommendations. The recommendation results can include suitable project opportunities, bidding opportunities, company opportunities, etc., which can be related stakeholders, unfinished projects, winning bid announcements and other information.
[0073] The method for recommending construction project information provided by the first aspect embodiment above can ensure the data quality and accuracy of the target construction project information by obtaining the construction project information to be analyzed, performing data cleaning on the construction project information to be analyzed, and obtaining the target construction project information. The target construction project information includes multiple target data entities. By constructing a project information association map based on multiple target data entities, the relationship between each target data entity can be reflected, which is convenient for subsequent in-depth data mining and analysis. By obtaining the user's subscription data and behavior data for construction projects, the user's preferences can be better reflected. The project information recommendation model is constructed according to the subscription data, behavior data and project information association map, which can improve the accuracy of the project information recommendation model. In response to the behavior instructions generated by the user visiting the target project page, the project information recommendation model is used to generate personalized recommendation results in real time, which can provide users with more accurate business opportunity recommendations. It can be understood that the present invention can enhance the fluidity and depth of data by constructing a project information association map using applied statistics and data analysis technology. At the same time, combined with the user's preferences, it can provide more reliable data support for the corresponding building materials companies to identify potential business opportunities and project needs, which is conducive to improving the company's operational efficiency.
[0074] It should be noted that the recommendation method of the embodiment of the present invention can improve the speed and efficiency of responding to the market by helping building materials companies to quickly identify potential business opportunities and project needs. It can also help building materials companies discover new market opportunities and expand their business scope and market coverage. By providing accurate building materials recommendations, companies can enhance customer relationships with construction companies and contractors. In addition, by analyzing construction project information, it can help building materials companies understand industry trends and promote product and service innovation.
[0075] It should be noted that applied statistics and data analysis techniques can be used to analyze a large amount of construction project information to be analyzed, enhance the fluidity of information data, and provide clear, comprehensive, and structured guidance and analysis data for the follow-up. Through the engineering information association map, it is possible to accurately judge the value of user business opportunities and deeply explore the data. Specifically, it can provide tabular, tree, and association-type aggregation analysis, so that users can grasp the value of business opportunities more accurately. In addition, the engineering information recommendation model can help users quickly identify and obtain key information, and quickly judge the follow-up value of the target company. The recommendation method of the embodiment of the present invention can help building materials companies enhance the fluidity of information data, expand users' in-depth use of data, enhance user stickiness, quickly judge the value of the target company's follow-up and grasp business opportunities, and improve overall operational efficiency and user experience.
[0076] In the above-mentioned construction project information recommendation method, in step S120, data cleaning is performed on the construction project information to be analyzed to obtain target construction project information, including but not limited to steps S210 to S250:
[0077] Step S210: traverse the construction project information to be analyzed, identify duplicate information and delete it to obtain first pre-processed information;
[0078] Step S220: performing format unification processing on the first preprocessing information according to a preset format rule to obtain second preprocessing information;
[0079] Step S230: performing missing value processing and outlier processing on the second preprocessing information to obtain third preprocessing information;
[0080] Step S240: performing desensitization processing on the third pre-processed information to obtain fourth pre-processed information;
[0081] Step S250: Perform data verification on the fourth preprocessing information according to preset verification rules to obtain target construction project information.
[0082] In this embodiment, the first pre-processed information is obtained by traversing the construction project information to be analyzed, identifying duplicate information and deleting it, that is, deduplication is performed to ensure the uniqueness of the data; then the first pre-processed information is formatted in a unified manner according to preset format rules to obtain second pre-processed information to ensure data consistency; then the second pre-processed information is missing value processed and outlier processed to obtain third pre-processed information; then the third pre-processed information is desensitized to obtain fourth pre-processed information to ensure protection of user privacy and data security; finally, the fourth pre-processed information is data verified according to preset verification rules to obtain target construction project information to ensure data accuracy and reliability.
[0083] Specifically, for step S210, for example, when the acquired construction project information includes project information p{name,...}, firstly, the abnormal characters in the project information are removed, including but not limited to: non-printing characters, special symbols, invisible characters, control characters, garbled characters, etc.; then, according to all the project information in the data set, a B+ search tree is constructed, and the project information can be quickly sorted and searched through the tree structure. In the process of traversing the data set, if duplicate project information is found, that is, the name field value is the same, only one piece of information is retained, and the rest of the duplicate information is deleted to ensure the uniqueness of the data.
[0084] For step S220, the data provided by different data sources may have inconsistent formats. For example, the date format may be "yyyy-mm-dd", "dd / mm / yyyy" or "mm / dd / yyyy". This step will use regular expressions and date parsing libraries to unify all date formats into the standard format of "yyyy-mm-dd".
[0085] For step S230, the specific method of missing value processing can be to choose to fill in missing values, delete information or use a model to predict missing values according to the nature and importance of the missing data. For outlier processing, obviously abnormal data, such as a negative project budget and a zero engineering area, will be automatically identified and marked for subsequent manual review and processing.
[0086] For step S240, since there may be data involving sensitive information, such as the company legal person ID number, bank account, etc., such data will be desensitized. The desensitization process can be performed using the following methods:
[0087] Replacement processing: Partially or completely replace sensitive data with a fixed string, such as "*" or "#";
[0088] Encryption processing: Use encryption algorithms to encrypt sensitive data, such as AES algorithm, RSA algorithm, etc.
[0089] For step S250, verification rules can be preset according to the business requirements of the construction project, and data verification can be performed on the fourth preprocessing information to ensure that the obtained target construction project information is consistent with the business requirements of the construction project.
[0090] In the above-mentioned construction project information recommendation method, step S130 constructs a project information association map based on multiple target data entities, including but not limited to steps S310 and S320:
[0091] Step S310: constructing entity attribute information according to each target data entity, and generating a corresponding entity information table, wherein the entity information table corresponds to the target data entity;
[0092] Step S320: Determine the relationship type between each target data entity, and construct an engineering information association map according to the relationship type.
[0093] In this embodiment, entity attribute information is constructed according to each target data entity, and a corresponding entity information table is generated, that is, the attributes of each entity are defined, and these attributes will be used as columns in the entity information table to reflect the relevant information of the target data entity. Then, the relationship type between each target data entity is determined, and an engineering information association map is constructed according to the relationship type. The engineering information association map can clearly show the connection and attributes between different target data entities.
[0094] Specifically, an engineering information association map is constructed through an entity-relationship diagram (ERD), which displays complex data relationships in a graphical way, making it more intuitive to understand and analyze data. The engineering information association map can be used to conduct in-depth data mining and analysis to discover hidden patterns and associations between data.
[0095] In one embodiment, the engineering information association graph includes detailed information of engineering projects, relationships between companies, contact information, etc. The relationship type between multiple target data entities can be one-to-one, one-to-many, or many-to-many. For example, engineering projects and companies: a project can be participated by multiple companies, and a company can participate in multiple projects; engineering projects and contacts: a project can have multiple contacts, and a contact can participate in multiple projects; companies and contacts: a company can have multiple contacts, and a contact can only belong to one company. In addition, for many-to-many relationships, an association table can be created to represent the relationship between target data entities.
[0096] In the above-mentioned construction project information recommendation method, in step S150, a project information recommendation model is constructed according to subscription data, behavior data and project information association graph, including but not limited to steps S410 to S460:
[0097] Step S410: Integrate subscription data, behavior data and project information association graph to generate a data source;
[0098] Step S420: extracting features from the data source and converting it into a feature data set, wherein the feature data set is key feature data that affects the construction project information;
[0099] Step S430: determining the data type of the feature data set, and determining the corresponding target recommendation algorithm according to the data type;
[0100] Step S440: training an engineering information recommendation model using a target recommendation algorithm and a feature data set;
[0101] Step S450: determining the performance evaluation index of the engineering information recommendation model;
[0102] Step S460: Optimize the parameters of the engineering information recommendation model according to the performance evaluation index to obtain a trained engineering information recommendation model.
[0103] In this embodiment, subscription data, behavior data and engineering information association graph are integrated to generate a unified data source for subsequent data processing; the data source is then subjected to feature extraction processing to be converted into a feature data set, i.e., key feature data affecting construction engineering information are extracted, and these key feature data are features useful for recommendation results; the data type of the feature data set is then determined, and the corresponding target recommendation algorithm is determined according to the data type; the target recommendation algorithm and the feature data set are then used to train the engineering information recommendation model, so that the output results of the engineering information recommendation model are more accurate; the performance evaluation indicators of the engineering information recommendation model are then determined, the performance evaluation indicators are used to evaluate the performance of the engineering information recommendation model, and the parameters of the engineering information recommendation model are optimized according to the performance evaluation indicators to obtain a trained engineering information recommendation model, thereby ensuring the accuracy and effectiveness of the model.
[0104] Specifically, for step S410, subscription data refers to the content that users actively choose to subscribe to or follow for construction projects. For example, subscription data includes information about users' attention to certain projects or companies. User behavior data refers to user operation records in corresponding construction project query services or applications. For example, behavior data includes user clicks, browsing, searching and other operation records on the corresponding construction project service system, while the project information association map includes detailed information on the project, the relationship between companies, contact information, etc. These data need to be cleaned, converted and standardized to ensure their accuracy and consistency.
[0105] For step S420, feature extraction engineering is performed on the basis of data integration to extract features that are useful for recommendation results. These features may include the user's historical behavior, interest preferences, activity, etc., and may also include the popularity, budget, progress, etc. of the project, as well as the company's credibility, strength, etc. Through feature extraction engineering, the original data can be converted into a feature set that is useful for the recommendation algorithm, that is, a feature data set, thereby improving the accuracy and efficiency of the recommendation results. Specifically, the original features are first encoded, and the text features are converted into digital features. Then, through standardization and normalization processing to eliminate the influence of different dimensions, multiple feature vectors are combined into new features through feature combination. Feature combination can be achieved through polynomial feature combination. For example, for two features x and y, feature data such as x^2, y^2, xy, etc. can be generated.
[0106] Specifically, the standardization processing formula is as follows:
[0107]
[0108] in, is the original eigenvalue, is the feature mean, is the characteristic standard deviation, is the standardized data.
[0109] Normalization is to scale the eigenvalues to a certain range. The normalization formula is as follows:
[0110]
[0111] in, is the original eigenvalue, is the minimum characteristic value, is the characteristic maximum value, is the normalized data.
[0112] For step S450 and step S460, the performance evaluation indicators of the engineering information recommendation model may include accuracy, recall rate, F1 score, etc. The performance of the engineering information recommendation model can be evaluated through the performance evaluation indicators, so as to optimize the parameters of the engineering information recommendation model so that the engineering information recommendation model can meet the requirements of various performance evaluation indicators, which is conducive to improving the generalization ability of the model.
[0113] In the above construction project information recommendation method, the data type of the feature data set is determined in step S430, and the corresponding target recommendation algorithm is determined according to the data type, including but not limited to steps S510 and S520:
[0114] Step S510: when it is determined that the feature data set is a first data type, the target recommendation algorithm is determined to be a similarity algorithm, wherein the first data type is used to reflect that the features of the feature data set are obvious;
[0115] Step S520: When it is determined that the feature data set is a second data type, the target recommendation algorithm is determined to be a deep learning recommendation algorithm, wherein the second data type is used to reflect that the data volume of the feature data set is large.
[0116] In this embodiment, the data type of the feature data set includes a first data type and a second data type. The first data type indicates that the feature data set has obvious characteristics, and the second data type indicates that the feature data set has a large amount of data. If the feature data set is the first data type, a similarity algorithm is selected, that is, a content-based recommendation algorithm. If the feature data set is the second data type, a deep learning recommendation algorithm is selected. By selecting appropriate target recommendation algorithms according to different data types, the accuracy of the recommendation results can be improved.
[0117] It should be noted that if the characteristics of the feature data set are obvious and easy to quantify, it is more effective to use a content-based recommendation algorithm. If the data volume of the feature data set is large and complex, a deep learning recommendation algorithm is more suitable.
[0118] In some embodiments, the content-based recommendation algorithm uses the jaccard similarity algorithm. For any two company sets, their feature value sets can be obtained, and then their jaccard similarity can be calculated. The specific formula is as follows:
[0119]
[0120] Among them, A and B represent the characteristic value sets of two companies respectively, and represents the size of the intersection of the two sets.
[0121] represents the size of the union of two sets, It is the similarity.
[0122] In some embodiments, the deep learning recommendation algorithm can be implemented in combination with the core concepts of reinforcement learning. For example, the user's historical behavior records, company, and contact features are represented as state vectors. , according to the current state vector Select the recommended action (For example, recommend a certain type of data), take action The feedback of the evaluation of the result after the action is determined, that is, the reward of the current action is determined , according to the reward , state vector ,action Optimize recommendation strategies so that deep learning recommendation algorithms can continuously learn and adapt to provide more personalized and accurate recommendations.
[0123] In the above-mentioned construction project information recommendation method, in step S160, in response to the behavior instruction generated by the user visiting the target project page, a recommendation result is generated according to the project information recommendation model, including but not limited to steps S610 to S630:
[0124] Step S610: in response to the action instruction generated by the user accessing the target project page, obtaining corresponding real-time project information;
[0125] Step S620: performing feature extraction processing on the real-time engineering information to obtain real-time feature data;
[0126] Step S630: input the real-time feature data into the engineering information recommendation model to obtain a recommendation result, wherein the recommendation result includes at least one of project business opportunity information, procurement business opportunity information, company business opportunity information, and a relationship map.
[0127] like Figure 2 As shown, in this embodiment, in response to the behavioral instructions generated by the user visiting the target project page, the corresponding real-time project information is obtained, and feature extraction processing is performed on the real-time project information to obtain real-time feature data. The real-time feature data is used to reflect the user's feature data. The real-time feature data is input into the project information recommendation model (i.e., the intelligent recommendation engine), and recommendation results corresponding to the user's behavioral characteristics can be obtained, thereby improving the accuracy of the recommendation.
[0128] It should be noted that when a user visits a target project page (such as a project details page, a company details page or a contact details page), relevant data needs to be obtained in real time, namely, real-time project information (such as detailed information on the project, relevant information on the company, relevant information on the contact, etc.). This data can be obtained from a database in real time through query or from other systems through an API interface.
[0129] After obtaining real-time engineering information, real-time feature calculation is required, such as calculating the jaccard similarity between the user and the current engineering project or company, or calculating the user's real-time behavior characteristics based on a deep learning model.
[0130] Specifically, the recommendation results may include project business opportunity information, procurement business opportunity information, company business opportunity information, relationship maps, etc., among which project business opportunity information includes relevant contacts, unfinished projects, contacts of unfinished projects, subordinate company projects and other information; procurement business opportunity information includes the number of public bidding announcements, the number of winning bid announcements, the number of exclusive procurement information and other information; company business opportunity information includes branch / subsidiary projects, the number of subordinate companies, the number of subordinate company projects and other information; the relationship map includes second-degree connections, public relations path display information, etc.
[0131] The above construction project information recommendation method also includes but is not limited to steps S710 to S730:
[0132] Step S710: Displaying the recommendation results in the user interaction interface;
[0133] Step S720: Obtaining user feedback information on the recommendation results;
[0134] Step S730: Optimize the engineering information recommendation model according to the feedback information.
[0135] In this embodiment, the generated recommendation results can be displayed to the user through the user interaction interface, for example, directly displayed on the web page of the user interaction interface or displayed through a pop-up window in the user interaction interface, and then the user's feedback information on the recommendation results in the user interaction interface is obtained, and the engineering information recommendation model is optimized according to the feedback information. By continuously optimizing and improving the recommendation model, the accuracy and efficiency of the recommendation can be improved, and the applicability to different users can also be improved.
[0136] It should be noted that the user interaction interface can provide interactive functions between users and recommendation results. For example, users can view detailed information of recommendation results, provide feedback and evaluation, or perform further operations on recommendation results. During the interaction between users and recommendation results, such as providing feedback and evaluation, user feedback information can be collected. This information can serve as an important basis for improving the recommendation model, thereby providing users with personalized business opportunity recommendation services.
[0137] In some embodiments, user feedback information on recommendation results can also be obtained by analyzing the user's operating behavior. For example, after displaying the recommendation results, feedback information can be obtained by analyzing the user's click actions, ignore actions, stay time and other data in the user interaction interface.
[0138] like Figure 3 As shown, in a second aspect, an embodiment of the present invention provides a construction engineering information recommendation system 300, comprising:
[0139] A first acquisition module 310 is used to acquire the construction project information to be analyzed;
[0140] A data cleaning module 320 is used to clean the construction project information to be analyzed to obtain target construction project information, wherein the target construction project information includes a plurality of target data entities;
[0141] The association map construction module 330 is used to construct an engineering information association map based on the multiple target data entities, wherein the engineering information association map is used to reflect the relationship between each of the target data entities;
[0142] The second acquisition module 340 is used to acquire the user's subscription data and behavior data for the construction project;
[0143] A recommendation model building module 350, configured to build a project information recommendation model based on the subscription data, the behavior data and the project information association graph;
[0144] The recommendation module 360 is used to generate a recommendation result according to the project information recommendation model in response to the behavior instruction generated by the user visiting the target project page.
[0145] According to the construction project information recommendation system provided by the embodiment of the present invention, by acquiring the construction project information to be analyzed, performing data cleaning on the construction project information to be analyzed, and obtaining the target construction project information, the data quality and accuracy of the target construction project information can be guaranteed. The target construction project information includes multiple target data entities. By constructing a project information association map based on multiple target data entities, the relationship between each target data entity can be reflected, which is convenient for subsequent in-depth data mining and analysis. By acquiring the user's subscription data and behavior data for construction projects, the user's preferences can be better reflected. The project information recommendation model is constructed according to the subscription data, behavior data and project information association map, which can improve the accuracy of the project information recommendation model. In response to the behavior instructions generated by the user visiting the target project page, the project information recommendation model is used to generate personalized recommendation results in real time, which can provide users with more accurate business opportunity recommendations. It can be understood that the present invention can enhance the fluidity and depth of data by constructing a project information association map using applied statistics and data analysis technology. At the same time, combined with the user's preferences, it can provide more reliable data support for the corresponding building materials companies to identify potential business opportunities and project needs, which is conducive to improving the operational efficiency of the company.
[0146] It should be noted that the specific implementation manner and corresponding technical effects of the construction project information recommendation system of the embodiment of the present invention may refer to the specific implementation manner and corresponding technical effects of the above-mentioned construction project information recommendation method.
[0147] like Figure 4 As shown, a third aspect of the present invention provides a computer device 400, including a memory 410, a processor 420, and a computer program stored in the memory 410 and executable on the processor 420, wherein the processor 420 and the memory 410 may be connected via a bus or other means. Figure 4 An example of connection via a bus is shown, and when the processor executes a computer program, the recommended method of the first aspect embodiment described above is implemented.
[0148] The computer device provided by the embodiment of the present invention can ensure the data quality and accuracy of the target construction project information by acquiring the construction project information to be analyzed, performing data cleaning on the construction project information to be analyzed, and obtaining the target construction project information. The target construction project information includes multiple target data entities. By constructing a project information association map based on multiple target data entities, the relationship between each target data entity can be reflected, which is convenient for subsequent in-depth data mining and analysis. By acquiring the user's subscription data and behavior data for the construction project, the user's preferences can be better reflected. The project information recommendation model is constructed according to the subscription data, behavior data and project information association map, which can improve the accuracy of the project information recommendation model. In response to the behavior instruction generated by the user visiting the target project page, the project information recommendation model is used to generate personalized recommendation results in real time, which can provide users with more accurate business opportunity recommendations. It can be understood that the present invention can enhance the fluidity and depth of data by constructing a project information association map using applied statistics and data analysis technology. At the same time, combined with the user's preferences, it can provide more reliable data support for the corresponding building materials companies to identify potential business opportunities and project needs, which is conducive to improving the operational efficiency of the company.
[0149] The fourth aspect embodiment of the present invention provides a computer-readable storage medium, which stores computer-executable instructions. The computer-executable instructions can be used to enable a computer to execute the recommended method of the first aspect embodiment as above. By acquiring the construction project information to be analyzed, performing data cleaning on the construction project information to be analyzed, and obtaining the target construction project information, the data quality and accuracy of the target construction project information can be guaranteed. The target construction project information includes multiple target data entities. By constructing a project information association map based on multiple target data entities, the relationship between each target data entity can be reflected, which is convenient for subsequent in-depth data mining and analysis. By acquiring the user's subscription data and behavior data for construction projects, the user's preferences can be better reflected. The project information recommendation model is constructed based on the subscription data, behavior data and project information association map, which can improve the accuracy of the project information recommendation model. In response to the behavior instructions generated by the user visiting the target project page, the project information recommendation model is used to generate personalized recommendation results in real time, which can provide users with more accurate business opportunity recommendations. It can be understood that the present invention can enhance the fluidity and depth of data by constructing a project information association map using applied statistics and data analysis technology. At the same time, combined with the user's preferences, it can provide more reliable data support for the corresponding building materials companies to identify potential business opportunities and project needs, which is conducive to improving the company's operational efficiency.
[0150] It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed methods above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium or a non-transitory medium and a communication medium or a temporary medium. As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information such as computer-readable instructions, data structures, program modules or other data. Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disk DVD or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.
[0151] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of ordinary technicians in the technical field without departing from the purpose of the present invention.
Claims
1. A method for recommending construction engineering information, characterized in that: include: Obtain the construction project information to be analyzed; Performing data cleaning on the construction project information to be analyzed to obtain target construction project information, wherein the target construction project information includes a plurality of target data entities; Constructing an engineering information association map based on the multiple target data entities, wherein the engineering information association map is used to reflect the relationship between each of the target data entities; The engineering information association map includes: detailed information of engineering projects, relationships between companies and contact information; Obtain user subscription data and behavior data for construction projects; Building a project information recommendation model according to the subscription data, the behavior data and the project information association graph; In response to a behavior instruction generated by a user visiting a target project page, generating a recommendation result according to the project information recommendation model; The step of constructing a project information recommendation model according to the subscription data, the behavior data and the project information association graph includes: Integrate the subscription data, the behavior data and the engineering information association graph to generate a data source; The subscription data includes: content that users actively choose to subscribe to or follow for construction projects; The user's behavior data includes: the user's click records, browsing records, and search records on the corresponding construction engineering service system; Performing feature extraction processing on the data source to convert it into a feature data set, wherein the feature data set is key feature data that affects construction engineering information; The key feature data include: user's historical behavior, interest preferences, activity, popularity of engineering projects, budget, progress, company's reputation and company's strength; Determine the data type of the feature data set, and determine the corresponding target recommendation algorithm according to the data type; Training an engineering information recommendation model using the target recommendation algorithm and the feature data set; Determining performance evaluation indicators of the engineering information recommendation model; Optimizing the parameters of the engineering information recommendation model according to the performance evaluation index to obtain a trained engineering information recommendation model; The determining the data type of the feature data set and determining the corresponding target recommendation algorithm according to the data type includes: When it is determined that the feature data set is a first data type, determining that the target recommendation algorithm is a similarity algorithm, wherein the first data type is used to reflect that the feature of the feature data set is obvious; The calculation formula of the similarity is as follows: Among them, A and B represent the characteristic value sets of two companies respectively. represents the size of the intersection of two sets, represents the size of the union of two sets, is the similarity; When it is determined that the feature data set is a second data type, determining that the target recommendation algorithm is a deep learning recommendation algorithm, wherein the second data type is used to reflect that the data volume of the feature data set is large; Wherein, the deep learning recommendation algorithm includes: Represent user historical behavior records, company, and contact features as state vectors , according to the current state vector Select the recommended action , take action The feedback of the evaluation of the result after the action is determined, that is, the reward of the current action is determined , according to the reward , state vector ,action Optimize recommendation strategy.
2. The recommendation method according to claim 1, characterized in that: The step of performing data cleaning on the construction project information to be analyzed to obtain target construction project information includes: Traversing the construction project information to be analyzed, identifying and deleting duplicate information, and obtaining first preprocessing information; Performing unified format processing on the first preprocessing information according to a preset format rule to obtain second preprocessing information; Performing missing value processing and outlier processing on the second preprocessed information to obtain third preprocessed information; Performing desensitization processing on the third preprocessing information to obtain fourth preprocessing information; The fourth preprocessing information is subjected to data verification according to preset verification rules to obtain target construction project information.
3. The recommendation method according to claim 1, characterized in that: The step of constructing an engineering information association graph based on the multiple target data entities includes: Constructing entity attribute information according to each of the target data entities, and generating a corresponding entity information table, wherein the entity information table corresponds to the target data entity; Determine the relationship type between each of the target data entities, and construct an engineering information association map based on the relationship type.
4. The recommendation method according to claim 1, characterized in that: The step of generating a recommendation result according to the project information recommendation model in response to the behavior instruction generated by the user visiting the target project page includes: In response to a behavior instruction generated by a user accessing a target project page, obtaining corresponding real-time project information; Performing feature extraction processing on the real-time engineering information to obtain real-time feature data; The real-time feature data is input into the engineering information recommendation model to obtain a recommendation result, wherein the recommendation result includes at least one of project business opportunity information, procurement business opportunity information, company business opportunity information, and a relationship map.
5. The recommendation method according to claim 1, characterized in that: Also includes: Displaying the recommendation results in a user interaction interface; Obtaining user feedback on the recommendation result; The engineering information recommendation model is optimized according to the feedback information.
6. A construction engineering information recommendation system, characterized in that: include: The first acquisition module is used to acquire the construction project information to be analyzed; A data cleaning module, used for performing data cleaning on the construction project information to be analyzed to obtain target construction project information, wherein the target construction project information includes a plurality of target data entities; An association map construction module, used to construct an engineering information association map based on the multiple target data entities, wherein the engineering information association map is used to reflect the relationship between each of the target data entities; The engineering information association map includes: detailed information of engineering projects, relationships between companies and contact information; The second acquisition module is used to acquire the user's subscription data and behavior data for the construction project; A recommendation model building module, used to build a project information recommendation model based on the subscription data, the behavior data and the project information association map; A recommendation module, configured to generate a recommendation result according to the project information recommendation model in response to a behavior instruction generated by a user visiting a target project page; The step of constructing a project information recommendation model according to the subscription data, the behavior data and the project information association graph includes: Integrate the subscription data, the behavior data and the engineering information association graph to generate a data source; The subscription data includes: the content that the user actively chooses to subscribe to or pay attention to for construction projects; the user's behavior data includes: The user's click records, browsing records, and search records on the corresponding construction engineering service system; Performing feature extraction processing on the data source to convert it into a feature data set, wherein the feature data set is key feature data that affects construction engineering information; The key feature data include: user's historical behavior, interest preferences, activity, popularity of engineering projects, budget, progress, company's reputation and company's strength; Determine the data type of the feature data set, and determine the corresponding target recommendation algorithm according to the data type; Training an engineering information recommendation model using the target recommendation algorithm and the feature data set; Determining performance evaluation indicators of the engineering information recommendation model; Optimizing the parameters of the engineering information recommendation model according to the performance evaluation index to obtain a trained engineering information recommendation model; The determining the data type of the feature data set and determining the corresponding target recommendation algorithm according to the data type includes: When it is determined that the feature data set is a first data type, determining that the target recommendation algorithm is a similarity algorithm, wherein the first data type is used to reflect that the feature of the feature data set is obvious; The calculation formula of the similarity is as follows: Among them, A and B represent the characteristic value sets of two companies respectively. represents the size of the intersection of two sets, represents the size of the union of two sets, is similarity; when it is determined that the feature data set is a second data type, the target recommendation algorithm is determined to be a deep learning recommendation algorithm, wherein the second data type is used to reflect that the data volume of the feature data set is large; Wherein, the deep learning recommendation algorithm includes: Represent user historical behavior records, company, and contact features as state vectors , according to the current state vector Select the recommended action , take action The feedback of the evaluation of the result after the action is determined, that is, the reward of the current action is determined , according to the reward , state vector ,action Optimize recommendation strategy.
7. A computer device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the recommendation method according to any one of claims 1 to 5 when executing the computer program.
8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the recommendation method according to any one of claims 1 to 5.
Citation Information
Patent Citations
Potential customer mining and customer model selection method based on hybrid recommendation algorithm
CN110930184A
Personalized learning resource recommendation method based on knowledge space representation
CN117216405A
Real-time film recommendation processing method and device based on knowledge graph
CN117972214A
Bidding and tendering data analysis method and system based on knowledge graph and medium
CN117974278A
