CN114911889A - Livable map recommendation method and device based on multi-source big data and machine learning - Google Patents

Livable map recommendation method and device based on multi-source big data and machine learning Download PDF

Info

Publication number
CN114911889A
CN114911889A CN202210579569.2A CN202210579569A CN114911889A CN 114911889 A CN114911889 A CN 114911889A CN 202210579569 A CN202210579569 A CN 202210579569A CN 114911889 A CN114911889 A CN 114911889A
Authority
CN
China
Prior art keywords
livable
behavior
map
user
preference
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210579569.2A
Other languages
Chinese (zh)
Inventor
仝德
龚咏喜
邱君丽
储君
潘向向
郑红霞
李汉廷
孙裔煜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Peking University Shenzhen Graduate School
Shenzhen Graduate School Harbin Institute of Technology
Original Assignee
Peking University Shenzhen Graduate School
Shenzhen Graduate School Harbin Institute of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Peking University Shenzhen Graduate School, Shenzhen Graduate School Harbin Institute of Technology filed Critical Peking University Shenzhen Graduate School
Priority to CN202210579569.2A priority Critical patent/CN114911889A/en
Publication of CN114911889A publication Critical patent/CN114911889A/en
Pending legal-status Critical Current

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
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • 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
    • G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • 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
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A30/00—Adapting or protecting infrastructure or their operation
    • Y02A30/60—Planning or developing urban green infrastructure

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Remote Sensing (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention relates to the technical field of big data analysis and processing, in particular to a livable map recommendation method and device based on multi-source big data and machine learning. The method comprises the steps of obtaining individual attributes and behavior preference data of a user, and constructing a behavior feature-livable preference mapping model; constructing an individualized livable map recommendation system in map software based on an SVM-GBRT model algorithm; and acquiring new user information, extracting user behavior characteristics to recommend a personalized livable map, and recommending a livable and feasible place evaluation result for the user. The personalized livable map is constructed and recommended, personalized livable evaluations oriented to different users are achieved, an iterative evolution model of individual behavior characteristics and subjective preferences is fused, personalization of the livable evaluations and diversification of service objects are achieved, and a livable map service mode is innovated.

Description

Livable map recommendation method and device based on multi-source big data and machine learning
Technical Field
The invention relates to the technical field of big data analysis and processing, in particular to a livable map recommendation method and device based on multi-source big data and machine learning.
Background
With the continuous development of economy and the continuous promotion of urbanization construction, the application of emerging data sources and technical means to assist the construction of livable cities becomes a new hot spot and trend.
However, the existing urban livable residence analysis and evaluation are carried out on the overall level of the city, and the individual livable residence requirements of residents need to be expanded and deepened urgently. Firstly, the existing urban livable evaluation system mostly aims at inter-city transverse comparison or urban self physical examination optimization, indexes capable of reflecting urban social and economic core competitiveness and overall image are screened from top to bottom, and the indexes reflecting individual living experience and inner feeling of residents from bottom to top are not sufficiently concerned. Secondly, with the development of internet and big data technology in recent years, although the selection and evaluation of livable indexes are no longer limited to statistical data with administrative regions as units, the space-time granularity of the data is still rough, and the difference of livable property at different times and different places in cities is ignored.
Secondly, people-oriented is a basic principle of habitability evaluation and planning construction, but the existing conception and methodology for building the habitability index system has little consideration on the feelings of residents or only considers the feelings of groups of residents to present evaluation results of one thousand people, the habitability preference and demand of the residents are ignored, and the evaluation results are difficult to reflect the individual difference. Moreover, the application scenes of the existing urban livable analysis and evaluation work are single, the urban livable analysis and evaluation method is mainly used for urban propaganda, inter-city comparison and evaluation, various application scenes such as enterprises and individuals are less considered, and the chances that the enterprises and residents participate in livable evaluation and enjoy livable evaluation results are few.
Based on the above, livable-oriented and smart city construction needs are urgently needed, a city personalized livable-oriented evaluation index system based on big data is constructed, a livable map construction and recommendation technology oriented to personalized needs is researched and developed, APP for urban residents to choose to make business references is developed, and therefore the livable-oriented and smart city evaluation personalization and service object diversification is achieved, and a livable map service mode is innovated.
Disclosure of Invention
The invention provides a personalized livable map recommendation method and device based on multi-source big data and a machine learning algorithm, which are used for establishing a behavior feature-livable preference model of urban residents by adopting methods such as machine learning and the like, so that the construction and recommendation of a personalized livable map are realized, and the technical problems are solved.
In order to achieve the above purpose, the embodiment of the present invention provides the following technical solutions:
in a first aspect, in an embodiment provided by the present invention, a livable map recommendation method based on multi-source big data and machine learning is provided, including the following steps:
acquiring user individual attributes and behavior preference data, wherein the user individual attributes comprise personal basic attributes, family attributes and behavior attributes;
analyzing the association among the individual attributes, the family attributes, the behavior attributes and the individual livable preferences based on a machine learning algorithm, and constructing a behavior feature-livable preference mapping model;
constructing an individualized livable map recommendation system based on a feature-livable preference mapping model in map software based on an SVM-GBRT model algorithm;
and acquiring new user information, and recommending personalized livable maps according to new user registration information and user behavior characteristics acquired by inputting the new user information into the characteristic-livable preference mapping model, so as to recommend livable and professional assessment results for the user.
As a further scheme of the invention, the behavior feature-livable preference mapping model is used for extracting effective features influencing the weight of the user livable index, and is also used for inputting the effective features to the feature-livable preference mapping model for continuous iteration and optimization according to new user information feedback data.
As a further scheme of the invention, before obtaining the individual attribute and behavior preference data of the user, the method also comprises the step of building a basic database based on the multisource spatio-temporal behavior big data, wherein the built basic database comprises the spatio-temporal behavior big data and the multisource basic data, and the method for building the basic database comprises the following steps:
and the resident individuals and the city basic components are taken as sensors to obtain basic dynamic information of the livable state of the city micro space and subjective and objective preference data of the resident individuals.
As a further aspect of the present invention, analyzing the association between the individual attribute, the family attribute, the behavior attribute and the individual livable preference based on a machine learning algorithm includes:
deducing the livable preference of the individual based on the basic attribute, the family attribute and the behavior attribute of the individual, analyzing the association among the individual attribute, the family attribute, the behavior attribute and the livable preference of the individual by adopting an XGboost algorithm, and constructing a behavior feature-livable preference model.
As a further scheme of the invention, a behavior characteristic-livable preference model is constructed, and the method further comprises the following steps: and continuously iterating and optimizing the feature-livable preference mapping model based on the livable preference fed back by the user.
As a further scheme of the invention, a personalized livable map recommendation system is constructed based on an SVM-GBRT model algorithm, and the personalized livable map recommendation system comprises the following steps: constructing a recommendation system based on an SVM-GBRT model based on data with geographic labels and implicit user interest;
the recommendation system consists of a matching process and a sorting process;
the matching process adopts a multivariate support vector machine model with OvR strategy to solve the behavior preference of the input user, and records and generates candidate objects;
in the sorting process, the candidate objects are scored and reordered by utilizing additional multi-source information and the SVM-GBRT model, and the evaluation results meeting the requirements in the suitable place of residence are recommended for the user.
As a further scheme of the invention, the livable map recommendation method based on the multi-source big data and the machine learning further comprises recommending a suitable livable and industriable assessment result for the user based on the input and output platform.
In a second aspect, in another embodiment provided by the present invention, a livable map recommendation device based on multi-source big data and machine learning is provided, and the livable map recommendation device based on multi-source big data and machine learning performs livable place recommendation by using the livable map recommendation method based on multi-source big data and machine learning; the livable map recommending device based on multi-source big data and machine learning comprises:
the data acquisition module is used for acquiring user individual attributes and behavior preference data, wherein the user individual attributes comprise personal basic attributes, family attributes and behavior attributes;
the model building module is used for analyzing the individual attributes, the family attributes and the association between the behavior attributes and the individual livable preferences based on a machine learning algorithm and building a behavior feature-livable preference mapping model;
the recommendation system module is used for constructing an individualized livable map recommendation system based on a feature-livable preference mapping model in map software based on an SVM-GBRT model algorithm;
and the map recommendation module is used for acquiring new user information, and recommending personalized livable maps according to new user registration information and user behavior characteristics acquired by inputting the new user information into the characteristic-livable preference mapping model, so as to recommend livable assessment results for the user.
In some embodiments provided by the present invention, the system further comprises a database construction module, configured to construct a basic database based on the multi-source spatio-temporal behavior big data, where the basic database includes the spatio-temporal behavior big data and the multi-source basic data.
In some embodiments provided by the invention, the system further comprises an input and output platform, wherein the input and output platform is used for visually displaying the evaluation result of the livable place recommended by the user, and the user is also used for adjusting the weight of the livable index according to the livable requirement so as to generate a livable map of the user.
In a third aspect, in yet another embodiment provided by the present invention, a computer device is provided, which includes a memory storing a computer program and a processor, which when loaded and executed, implements the steps of a livable map recommendation method based on multi-source big data and machine learning.
In a fourth aspect, in a further embodiment provided by the present invention, a storage medium is provided, which stores a computer program that is loaded by a processor and executed to implement the steps of the livable map recommendation method based on multi-source big data and machine learning.
The technical scheme provided by the invention has the following beneficial effects:
according to the livable map recommendation method and device based on the multi-source big data and the machine learning, the behavior feature perception, livable preference recognition and personalized livable map construction and recommendation based on the social perception big data and the machine learning are realized, personalized livable evaluation for different users is realized, an iterative evolution model integrating individual behavior features and subjective preference is realized, the personalization of livable evaluation and the diversification of service objects are realized, and a livable map service mode is innovated.
These and other aspects of the invention are apparent from and will be elucidated with reference to the embodiments described hereinafter. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention. In the drawings:
fig. 1 is a flowchart of a livable map recommendation method based on multi-source big data and machine learning according to an embodiment of the present invention.
Fig. 2 is a schematic frame diagram of a livingmap recommendation method based on multi-source big data and machine learning according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of user objective preference data in a livable map recommendation method based on multi-source big data and machine learning according to another embodiment of the invention.
Fig. 4 is a system block diagram of a livable map recommendation device based on multi-source big data and machine learning according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In some of the flows described in the present specification and claims and in the above figures, a number of operations are included that occur in a particular order, but it should be clearly understood that these operations may be performed out of order or in parallel as they occur herein, with the order of the operations being indicated as 101, 102, etc. merely to distinguish between the various operations, and the order of the operations by themselves does not represent any order of performance. Additionally, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first", "second", etc. in this document are used for distinguishing different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit the types of "first" and "second".
The technical solutions in the exemplary embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the exemplary embodiments of the present invention, and it is apparent that the described exemplary embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The current empirical paradigm for urban livable planning is based on statistical data and implements spatial layout according to various normative criteria and objective standardized indicators. With the transformation of planning targets to the high-quality connotation such as containment, wisdom, fineness and the like, a living evaluation system of 'one whole city, one whole year, one thousand people' developed on the basis of the whole city level is more suitable for the research and comparison on the macro level, and the refining and individualization requirements of microscopic scale, space evaluation and planning in the city are difficult to meet.
Aiming at the problems that the existing livable city evaluation system is more suitable for comparison on a macroscopic level, residents are homogenized, personalized requirements behind a single number are ignored, the fine requirements of city management are difficult to meet, service objects are single, and application scenes need to be expanded urgently, the livable map recommendation method and the livable map recommendation device based on multi-source big data and machine learning are used for establishing a city resident behavior characteristic-livable preference model by adopting methods such as machine learning and the like, and construction and recommendation of the personalized livable map are realized, so that the technical problems are solved.
Specifically, the embodiments of the present application will be further explained below with reference to the drawings.
Example 1
Referring to fig. 1 and 2, an embodiment of the present invention provides a livable map recommendation method based on multi-source big data and machine learning, including steps S10-S40;
s10, acquiring user individual attributes and behavior preference data, wherein the user individual attributes comprise personal basic attributes, family attributes and behavior attributes;
in the embodiment of the application, the individual attribute and behavior preference data of residents are obtained, the individual attribute and the behavior preference data comprise individual basic attributes such as sex, age and occupation, family attributes such as family number, old people and children, behavior attributes such as residence, employment, trip and leisure, and effective characteristics influencing the livable index weight of the residents are extracted.
S20, analyzing the individual attribute, the family attribute and the association between the behavior attribute and the individual livable preference based on a machine learning algorithm, and constructing a behavior feature-livable preference mapping model;
s30, constructing a personalized livable map recommendation system based on a feature-livable preference mapping model in map software based on an SVM-GBRT model algorithm;
s40, obtaining new user information, and recommending personalized livable maps according to the new user registration information and the user behavior characteristics obtained by inputting the new user information into the characteristic-livable preference mapping model, so as to recommend livable place evaluation results for the user.
In the embodiment, personalized livable analysis and service innovation of social perception theoretical methods such as space-time behavior big data and machine learning are focused, an urban personalized livable evaluation index system and a calculation method are built, and personalized demand oriented livable map building and recommendation technologies are developed and applied to the fields of livable urban planning, characteristic space building, space customization, user locking and the like, so that a technical achievement system with independent intellectual property rights and wide market popularization prospects is formed.
In this embodiment, the behavior feature-livable preference mapping model is used for extracting effective features affecting the weight of the user livable index, and is further used for inputting the effective features to the feature-livable preference mapping model for continuous iteration and optimization according to new user information feedback data.
In order to enrich the urban livable evaluation method system, in the embodiment of the invention, before the user individual attribute and behavior preference data are obtained, a basic database is built on the basis of multi-source space-time behavior big data. The constructed basic database comprises space-time behavior big data and multi-source basic data, and the method for constructing the basic database comprises the following steps: the basic dynamic information of the livable state of the urban micro space and subjective and objective preference data of the resident are obtained by taking the resident and the urban basic components as sensors, and the objective preference data of the user includes social networks, online shopping consumption, mobile activities, emotional cognition and the like, for example, as shown in fig. 3.
When subjective preference data of a user are collected, active big data of the user about a livable city are collected through a mobile phone APP form, and evaluation data of satisfaction or acceptance degrees of residents to the whole city, different areas of the city, streets and the like are obtained.
In the embodiment, in the aspect of data mining, by means of the internet of things and the WEB2.0 technology, the basic dynamic information of the livable state of the urban micro space and the subjective and objective preference data of the resident are obtained by taking the resident and the basic parts of the urban as sensors, so that the evaluation of urban livable state has objectivity, dynamics and individuation.
Therefore, a database comprising the space-time behavior big data and the multi-source basic data is constructed: the method comprises the steps of firstly, using space-time behavior big data such as mobile phone signaling, taxi tracks, bus card swiping, social media and environment real-time monitoring; and secondly, multi-source basic data such as land utilization, road networks, public supporting facilities and the like are integrated.
The urban livable single index evaluation technology based on multi-source space-time behavior big data is based on space-time behavior big data such as mobile phone signaling, taxi tracks, bus swiping cards and social media, and adopts methods such as an improved two-step mobile search method, road network matching and spatial interpolation to calculate and standardize single livable indexes from multiple dimensions in order to construct a general urban livable map.
Illustratively, a grid of 500 × 500 meters is used as a basic space unit, a single element evaluation result is assigned according to 0-100 points, multi-source data standardization is completed, and multi-element index fusion is performed through space superposition. And developing personal figures of the client by utilizing information such as a user space movement track, consumption, a social network and the like, constructing a livable evaluation system aiming at personal attributes of the client, and carrying out visualization of urban livable space evaluation and dynamic change and display of a map.
In this embodiment, analyzing the association between the individual attribute, the family attribute, the behavior attribute and the individual livable preference based on the machine learning algorithm includes: deducing the livable preference of the individual based on the basic attribute, the family attribute and the behavior attribute of the individual, analyzing the association among the individual attribute, the family attribute, the behavior attribute and the individual livable preference by adopting an XGboost algorithm, and constructing a behavior feature-livable preference model.
In this embodiment, constructing the behavior feature-livable preference model further includes: continuously iterating and optimizing the feature-livable preference mapping model based on the livable preference fed back by the user, and improving the precision of the behavior feature-livable preference model.
When the resident attribute features are extracted, the feature extraction is used as an important component of a machine learning method and is an important premise for ensuring the accuracy of a machine model. In the embodiment of the application, a convolutional neural network is used for extracting features, the convolutional neural network mainly comprises three structures of convolution, activation and pooling, the convolution process is to move a convolution kernel in a two-dimensional space, and the sum of all weights of the convolution kernel and corresponding weights on input samples of the convolution kernel is calculated and can be represented as follows:
Figure 467202DEST_PATH_IMAGE001
where the convolution kernel weights w, the value is v. Then an activation function is introduced:
Figure 914233DEST_PATH_IMAGE002
wherein the content of the first and second substances,bin order to be offset,h() For an activation function, p is the true distribution of the activation function and q is the non-true distribution of the activation function. The present study used a linear rectification unit function (ReLU):
Figure 461889DEST_PATH_IMAGE003
. And then, down-sampling operation is further performed through pooling operation, so that the feature space dimension is reduced.
And performing convolution, planning and pooling iterative operation until the result feature vector is optimal. And finally obtaining the attribute characteristics of the residents according with the current livable indexes.
When the individual behavior feature-livable preference model of residents is constructed based on the XGboost algorithm, the individual behavior feature-livable preference is constructed by the XGboost algorithm on the basis of determining the most resident feature vector of each livable index. XGboost is used as one of Boosting algorithms, and XGboost is used for establishing a behavior characteristic-livable preference model.
Exemplary, for example, an XGBoost model composed of K CART regression trees using a known model:
Figure 428708DEST_PATH_IMAGE004
wherein
Figure 934775DEST_PATH_IMAGE005
Is the total number of the trees,
Figure 809715DEST_PATH_IMAGE006
is shown as
Figure 161062DEST_PATH_IMAGE007
The number of the tree is one, and the tree is one,
Figure 982387DEST_PATH_IMAGE008
representing a sample
Figure 643045DEST_PATH_IMAGE009
The predicted result of (1). The objective function is:
Figure 549821DEST_PATH_IMAGE010
the loss function is expressed as:
Figure 704859DEST_PATH_IMAGE011
wherein
Figure 380691DEST_PATH_IMAGE012
Is a sample
Figure 212249DEST_PATH_IMAGE013
The error of the training of (a) is,
Figure 606321DEST_PATH_IMAGE014
representing a sample
Figure 299471DEST_PATH_IMAGE013
As a result of the training of (a),
Figure 829809DEST_PATH_IMAGE015
is shown as
Figure 832269DEST_PATH_IMAGE016
Regular terms of the tree.
And through continuous iterative training, the loss is gradually reduced, and a better model is finally obtained.
In this embodiment, a personalized livable map recommendation system is constructed based on an SVM-GBRT model algorithm, and the method includes: and constructing a recommendation system based on the SVM-GBRT model based on social media data with geographic labels, behavior data and other data which imply user interest. The recommendation system is composed of a matching process and a sorting process.
The matching process adopts a multivariate support vector machine model with OvR strategy to solve the behavior preference of the input user, and records and generates candidate objects; in the sorting process, the candidate objects are scored and reordered by utilizing additional multi-source information and the SVM-GBRT model, and the evaluation results meeting the requirements in the suitable place of residence are recommended for the user.
A model-based recommendation system is adopted and consists of two stages of matching and ranking contents. In the matching process, the multivariate support vector machine model (the multiclass SVM model) adopting OvR (the one-vs. -rest) strategy is adopted to generate candidate objects by inputting the travel history of the user. In the sorting process, additional multi-source information is utilized, a GBRT gradient enhanced regression model is applied to score candidates, and the candidate list is reordered.
In an embodiment of the application, the livable map recommendation method based on the multi-source big data and the machine learning further comprises the step of recommending a suitable livable-livable place evaluation result for the user based on the input and output platform. On the premise that the single livable evaluation index and the comprehensive livable index algorithm of the basic database are clear, the dynamic livable map can be displayed through the input and output platform, and the personalized livable map can be generated according to the personal index weight.
It should be understood that although the steps are described above in a certain order, the steps are not necessarily performed in the order described. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, some steps of the present embodiment may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the steps or stages is not necessarily sequential, but may be performed alternately or in turns with other steps or at least a part of the steps or stages in other steps.
Example 2
Referring to fig. 4, in an embodiment of the present invention, a livable map recommendation device based on multi-source big data and machine learning is further provided, and the livable map recommendation device based on multi-source big data and machine learning performs livable place recommendation by using the livable map recommendation method based on multi-source big data and machine learning, and includes a data acquisition module 100, a model construction module 200, a recommendation system module 300, a map recommendation module 400, a database construction module 500, and an input/output platform 600.
The data acquisition module 100 is configured to acquire user individual attributes and behavior preference data, where the user individual attributes include a personal basic attribute, a family attribute, and a behavior attribute;
the model construction module 200 is configured to analyze the association between the individual attribute, the family attribute, the behavior attribute and the individual livable preference based on a machine learning algorithm, and construct a behavior feature-livable preference mapping model;
the recommendation system module 300 is used for constructing an individualized livable map recommendation system based on a feature-livable preference mapping model in map software based on an SVM-GBRT model algorithm;
the map recommendation module 400 is configured to obtain new user information, and perform personalized livable map recommendation according to new user registration information and user behavior characteristics obtained by inputting the new user information into the characteristic-livable preference mapping model, so as to recommend a livable place evaluation result for the user.
The database construction module 500 is configured to construct a basic database based on the multi-source temporal-spatial behavior big data, where the basic database includes the temporal-spatial behavior big data and the multi-source basic data.
The input and output platform 600 is used for visually displaying the evaluation result of the livable place recommended by the user, and is also used for adjusting the weight of the livable index according to the livable demand so as to generate a livable map of the user.
It should be noted that, when the livable map recommendation device based on the multi-source big data and the machine learning is executed, the steps of the livable map recommendation method based on the multi-source big data and the machine learning are adopted, so that the operation process of the livable map recommendation device based on the multi-source big data and the machine learning in this embodiment is not described in detail.
Example 3
In an embodiment, there is further provided, in an embodiment of the present invention, a computer device, including at least one processor, and a memory communicatively connected to the at least one processor, where the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to cause the at least one processor to execute the method for recommending a livingwap based on multi-source big data and machine learning, and the processor executes the instructions to implement the steps in the method embodiments:
acquiring user individual attributes and behavior preference data, wherein the user individual attributes comprise personal basic attributes, family attributes and behavior attributes;
analyzing the association among the individual attributes, the family attributes, the behavior attributes and the individual livable preferences based on a machine learning algorithm, and constructing a behavior feature-livable preference mapping model;
constructing an individualized livable map recommendation system based on a feature-livable preference mapping model in map software based on an SVM-GBRT model algorithm;
and acquiring new user information, and recommending personalized livable maps according to new user registration information and user behavior characteristics acquired by inputting the new user information into the characteristic-livable preference mapping model, so as to recommend livable and professional assessment results for the user.
Example 4
In an embodiment of the present invention, there is further provided a storage medium having stored thereon a computer program which, when executed by a processor, implements the steps in method embodiments 1 described above:
acquiring user individual attributes and behavior preference data, wherein the user individual attributes comprise personal basic attributes, family attributes and behavior attributes;
analyzing the association among the individual attributes, the family attributes, the behavior attributes and the individual livable preferences based on a machine learning algorithm, and constructing a behavior feature-livable preference mapping model;
constructing an individualized livable map recommendation system based on a feature-livable preference mapping model in map software based on an SVM-GBRT model algorithm;
and acquiring new user information, and recommending personalized livable maps according to new user registration information and user behavior characteristics acquired by inputting the new user information into the characteristic-livable preference mapping model, so as to recommend livable and professional assessment results for the user.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory.
In conclusion, the livable map recommendation method and device based on the multi-source big data and the machine learning provided by the invention are based on the behavior feature perception, livable preference recognition and personalized livable map construction and recommendation of the social perception big data and the machine learning, so that personalized livable evaluation for different users is realized, an iterative evolution model integrating individual behavior features and subjective preference is integrated, personalization of livable evaluation and diversification of service objects are realized, and a livable map service mode is innovated.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.

Claims (10)

1. A livable map recommendation method based on multi-source big data and machine learning is characterized by comprising the following steps:
acquiring user individual attributes and behavior preference data, wherein the user individual attributes comprise personal basic attributes, family attributes and behavior attributes;
analyzing the association among the individual attributes, the family attributes, the behavior attributes and the individual livable preferences based on a machine learning algorithm, and constructing a behavior feature-livable preference mapping model;
constructing an individualized livable map recommendation system based on a feature-livable preference mapping model in map software based on an SVM-GBRT model algorithm;
and acquiring new user information, and recommending personalized livable maps according to new user registration information and user behavior characteristics acquired by inputting the new user information into the characteristic-livable preference mapping model, so as to recommend livable and professional assessment results for the user.
2. The multi-source big data and machine learning-based livable map recommendation method of claim 1, wherein the behavior feature-livable preference mapping model is used for extracting effective features influencing the weight of the livable indicators of the user, and is further used for inputting the effective features into the feature-livable preference mapping model for continuous iteration and optimization according to new user information feedback data.
3. The multi-source big data and machine learning-based livable map recommendation method of claim 2, wherein constructing a behavior feature-livable preference model further comprises: and continuously iterating and optimizing the feature-livable preference mapping model based on the livable preference fed back by the user.
4. The livingMaps recommendation method based on multi-source big data and machine learning of claim 1, further comprising building a basic database based on the multi-source spatiotemporal behavior big data before obtaining the individual attributes and behavior preference data of the user, wherein the built basic database comprises the spatiotemporal behavior big data and the multi-source basic data, and the method for building the basic database comprises the following steps:
and the basic dynamic information of the livable state of the urban micro space and the subjective and objective preference data of the resident individuals are obtained by taking the resident individuals and the urban basic components as sensors.
5. The livable map recommendation method based on multi-source big data and machine learning of claim 4, wherein analyzing the association between the individual attributes, the family attributes, the behavior attributes and the individual livable preferences based on a machine learning algorithm comprises:
deducing the livable preference of the individual based on the basic attribute, the family attribute and the behavior attribute of the individual, analyzing the association among the individual attribute, the family attribute, the behavior attribute and the individual livable preference by adopting an XGboost algorithm, and constructing a behavior feature-livable preference model.
6. The livable map recommendation method based on multi-source big data and machine learning of claim 1 or 5, wherein a personalized livable map recommendation system is constructed based on SVM-GBRT model algorithm, and comprises the following steps: constructing a recommendation system based on an SVM-GBRT model based on data with geographic labels and implicit user interest;
the recommendation system consists of a matching process and a sorting process;
the matching process adopts a multivariate support vector machine model with OvR strategy to solve the behavior preference of the input user, and records and generates candidate objects;
in the sorting process, the candidate objects are scored and reordered by utilizing additional multi-source information and the SVM-GBRT model, and the evaluation results meeting the requirements in the suitable place of residence are recommended for the user.
7. The multi-source big data and machine learning-based livable map recommendation method of claim 1, further comprising recommending a suitable livable place evaluation result for the user based on the input-output platform.
8. A livable map recommending device based on multi-source big data and machine learning, characterized in that the livable map recommending method based on multi-source big data and machine learning of any one of claims 1-7 is adopted for livable place recommendation; the livable map recommending device based on multi-source big data and machine learning comprises:
the data acquisition module is used for acquiring user individual attributes and behavior preference data, wherein the user individual attributes comprise personal basic attributes, family attributes and behavior attributes;
the model building module is used for analyzing the individual attributes, the family attributes and the association between the behavior attributes and the individual livable preferences based on a machine learning algorithm and building a behavior feature-livable preference mapping model;
the recommendation system module is used for constructing an individualized livable map recommendation system based on a feature-livable preference mapping model in map software based on an SVM-GBRT model algorithm;
and the map recommendation module is used for acquiring new user information, and recommending personalized livable maps according to new user registration information and user behavior characteristics acquired by inputting the new user information into the characteristic-livable preference mapping model, so as to recommend livable assessment results for the user.
9. The livingMaps recommendation device based on multisource big data and machine learning of claim 8, further comprising a database construction module for constructing a base database based on multisource spatiotemporal behavior big data, the base database comprising spatiotemporal behavior big data and multisource base data.
10. The livable map recommending device based on multi-source big data and machine learning of claim 9, further comprising an input-output platform for visually displaying the livable place assessment result recommended by the user, and further for the user to adjust the weight of the livable index according to the livable demand to generate the livable map of the user.
CN202210579569.2A 2022-05-26 2022-05-26 Livable map recommendation method and device based on multi-source big data and machine learning Pending CN114911889A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210579569.2A CN114911889A (en) 2022-05-26 2022-05-26 Livable map recommendation method and device based on multi-source big data and machine learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210579569.2A CN114911889A (en) 2022-05-26 2022-05-26 Livable map recommendation method and device based on multi-source big data and machine learning

Publications (1)

Publication Number Publication Date
CN114911889A true CN114911889A (en) 2022-08-16

Family

ID=82768497

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210579569.2A Pending CN114911889A (en) 2022-05-26 2022-05-26 Livable map recommendation method and device based on multi-source big data and machine learning

Country Status (1)

Country Link
CN (1) CN114911889A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115796633A (en) * 2022-09-30 2023-03-14 北京大学深圳研究生院 Urban rural land utilization comprehensive improvement performance assessment method and system

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115796633A (en) * 2022-09-30 2023-03-14 北京大学深圳研究生院 Urban rural land utilization comprehensive improvement performance assessment method and system
CN115796633B (en) * 2022-09-30 2023-08-08 北京大学深圳研究生院 Urban village land utilization comprehensive renovation performance evaluation method and system

Similar Documents

Publication Publication Date Title
Song et al. A review of research on tourism demand forecasting: Launching the Annals of Tourism Research Curated Collection on tourism demand forecasting
He et al. A classification of public transit users with smart card data based on time series distance metrics and a hierarchical clustering method
Kadoić et al. Integrating the DEMATEL with the analytic network process for effective decision-making
Li et al. Statistical analysis of tourist flow in tourist spots based on big data platform and DA-HKRVM algorithms
Dadkhah et al. Corporate foresight: developing a process model
Chang et al. Application of neural networks and Kano’s method to content recommendation in web personalization
Hanafi et al. Modelling competitive advantage of nation: a literature review
Fernandez et al. Core: A decision support system for regional competitiveness analysis based on multi-criteria sorting
Casillo et al. Context-aware recommender systems and cultural heritage: a survey
Gomez-Vega et al. Clustering and country destination performance at a global scale: Determining factors of tourism competitiveness
Pantano et al. New technology and tourism industry innovation: evidence from audio-visual patented technologies
Park et al. Two-stage technology opportunity discovery for firm-level decision making: GCN-based link-prediction approach
Deng et al. Tourism destination preference prediction based on edge computing
Santos et al. A methodology for the identification of strategic technological competences: An application in the sheet metal equipment industry
CN114911889A (en) Livable map recommendation method and device based on multi-source big data and machine learning
Mohammed et al. A decision modeling approach for smart e-tourism data management applications based on spherical fuzzy rough environment
Zhang et al. [Retracted] Application and Analysis of Artificial Intelligence in College Students’ Career Planning and Employment and Entrepreneurship Information Recommendation
Alnusairat et al. Information system, geography, information management system and tourism planning: a geographical perspective from Malaysia
Lai et al. Explaining a century of Swiss regional development by deep learning and SHAP values
Mirabella et al. The population equivalent as a novel approach for life cycle assessment of cities and inter-city comparisons
Imani et al. A multiple-discrete approach for examining vehicle type use for daily activity participation decisions
Noorian A BERT-based sequential POI recommender system in social media
Bi et al. Personalized recommendation of rural tourism based on traffic classification and user data analysis
Bai et al. Siamese-like convolutional neural network for fine-grained income estimation of developed economies
Anamisa et al. K-Nearest Neighbors Method for Recommendation System in Bangkalan’s Tourism

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