A Review of Multi-Source Data Fusion and Analysis Algorithms in Smart City Construction: Facilitating Real Estate Management and Urban Optimization

Abstract

In the context of the booming construction of smart cities, multi-source data fusion and analysis algorithms play a key role in optimizing real estate management and improving urban efficiency. In this review, we comprehensively and systematically review the relevant algorithms, covering the types, characteristics, fusion techniques, analysis algorithms, and their synergies of multi-source data. We found that multi-source data, including sensors, social media, citizen feedback, and GIS data, face challenges such as data quality and privacy security when being fused. Data fusion algorithms are diverse and have their own advantages and disadvantages. Data analysis algorithms help urban management in areas such as spatial analysis and deep learning. Algorithm collaboration can improve decision-making accuracy and efficiency and promote the rational allocation of urban resources. In the future, algorithm development will focus on data quality, real-time, deep mining, interdisciplinary research, privacy protection, and collaborative application expansion, providing strong support for the sustainable development of smart cities.

1. Introduction

In the context of the rapid development of smart cities, multi-source data fusion and analysis algorithms play a vital role in optimizing real estate management, which encompasses activities such as property valuation, market analysis, investment strategy, regulatory compliance, and urban efficiency. The urbanization process brings challenges such as population growth, resource shortages, and environmental pollution, and smart city construction is considered to be an effective way to solve these problems [1]. At present, urban real estate management is restricted by information asymmetry, low resource utilization efficiency, and unreasonable decision-making, which affects its sustainable development [2]. Multi-source data fusion technology can integrate different types of data from sensors, social media, public feedback, and GIS to achieve comprehensive and real-time information perception, thereby providing a scientific decision-making basis for real estate management [3]. Through deep mining of data, analysis algorithms can effectively provide urban managers with support for evaluation, planning, and supervision, improve the efficiency of real estate management, optimize urban spatial layout, and promote the development of cities towards sustainability and intelligence [4]. A large number of studies have shown that multi-source data fusion and analysis algorithms have many key meanings in smart cities. First, multi-source data fusion technology enables governments and enterprises to better control real estate information and improve the efficiency of management and decision-making. Secondly, through data analysis, urban planners can deeply explore the laws of urban space utilization, thereby achieving more efficient resource allocation [5]. Furthermore, these technologies are capable of identifying a range of potential issues in urban development, such as traffic congestion, inadequate housing, and inadequate public services. For instance, in the case of traffic congestion, smart data analysis can help in optimizing traffic flow by predicting peak hours and suggesting alternative routes. In terms of housing, multi-source data can inform policymakers about the demand and supply dynamics, enabling them to develop strategies to address housing shortages. Additionally, these technologies can provide insights into the distribution and quality of public services, aiding in the equitable allocation of resources. By uncovering such challenges, these technologies offer policymakers targeted solutions, thereby fostering the green and sustainable development of cities [6]. The quality of life of residents is also improved by these technologies, and public service facilities can be optimized by better understanding residents’ needs [7]. In addition, research and development of multi-source data fusion and analysis algorithms will help promote the innovation of related technologies and provide continuous technical support for the construction of smart cities [8]. This study provides a theoretical basis and practical guidance for solving the challenges of urban real estate management, optimizing spatial layout, and enhancing sustainable development capabilities.

In this literature review, we review the application of multi-source data fusion and analysis algorithms in smart city construction to assist real estate management and urban optimization. In Section 2, we discuss in detail the multi-source data in smart city construction, including various data sources such as sensors, social media, citizen feedback, and GIS data. This section focuses on analyzing the sources, characteristics, and challenges faced in practical applications of these data, which lays the foundation for data fusion and analysis in subsequent sections. In Section 3, we discuss the principles and methods of different data fusion technologies, including data preprocessing, data matching and association, and various data fusion algorithms. This section emphasizes the role of data fusion in improving real estate management efficiency and urban optimization. In Section 4, we turn to reviewing the application of data analysis algorithms in urban real estate management, focusing on the application scenarios and effects of spatial data analysis and deep-learning algorithms. This section analyzes how spatial statistical analysis and geospatial data mining algorithms can assist in assessing real estate value and optimizing urban planning and discusses the role of deep learning in real estate data mining. In Section 5, we reviewed the synergy between algorithms and their application effects in smart city optimization and explained how to improve the efficiency of real estate management and urban optimization by coordinating different data fusion and analysis algorithms. We emphasized that algorithm synergy improves the accuracy and efficiency of data processing. We also promote the rational allocation of urban resources and propose the challenges and future development trends of algorithm synergy. Finally, in Section 6, we draw the final conclusion of this review.

2. Methods

2.1. Literature Sources

In this study, in order to comprehensively and deeply explore the application of multi-source data fusion and analysis algorithms in smart city construction, we have extensively collected literature from multiple channels to ensure the comprehensiveness and reliability of the research.

2.1.1. Academic Databases

Academic databases are an important way to obtain high-quality research literature. We mainly used well-known databases such as Web of Science, IEEE Xplore, and Elsevier ScienceDirect. These databases cover many disciplines, such as computer science, geography, urban planning, information science, etc., which are highly consistent with the multidisciplinary nature of this study. By using relevant keywords (such as “multi-source data fusion”, “smart city”, “real estate management”, “data analysis algorithm”, etc.) in these databases for combined retrieval, we obtained a large number of peer-reviewed academic papers, including research reports and review articles. These documents not only provide a rich theoretical basis but also introduce in detail the cases and effect evaluation of various algorithms in practical applications, which provides a solid basis for our in-depth understanding of the role of multi-source data fusion and analysis algorithms in smart city construction.

2.1.2. Government and Research Institution Reports

Reports and publications issued by government agencies and professional research institutions are also an important part of our data sources. For example, the guidance documents and practical case analysis on smart city construction issued by the Ministry of Housing and Urban-Rural Development provide us with macro-level information on the policy orientation, development planning, and data application needs of smart city construction in actual projects. At the same time, special research reports on smart city construction in local areas issued by urban planning institutes, geographic information research centers, and other institutions contain rich, localized data and practical experience, which helps us to deeply analyze the characteristics and differences in multi-source data fusion and analysis algorithm application in different regions. These reports are usually based on actual surveys and project implementations and have strong practical guiding significance, which can help us better grasp the application scenarios and challenges of algorithms in actual smart city construction.

2.1.3. Industry Publications and Conference Proceedings

Industry publications such as “Journal of Information Science” and “Urban Planning International” focus on the latest technological advances, application case sharing, and industry dynamic analysis in the field of smart city construction. By reading these journals, we have obtained a lot of timely and practical information and learned about the latest technology trends and application hotspots in the industry, such as the application innovation of emerging data fusion algorithms and data analysis tools in various fields of smart cities (such as transportation, energy, environment, etc.). In addition, the proceedings of relevant academic conferences are also valuable resources for obtaining cutting-edge research results. These conferences bring together experts, scholars, and industry practitioners from around the world. The latest research results and practical experience they share at conferences often represent the cutting-edge dynamics in the field. Conference papers cover all aspects, from theoretical research to practical applications, providing us with rich research ideas and innovative methods and helping us to keep up with the latest research directions and technological breakthroughs in the field.

2.2. Literature Selection Criteria

In order to ensure that the selected literature can accurately serve the purpose of this study, that is, to deeply explore the role of multi-source data fusion and analysis algorithms in promoting real estate management and urban optimization in smart city construction, we have formulated strict and systematic literature selection criteria.

2.2.1. Relevance Principle

The literature must be closely centered on the core background of smart city construction and focus on the application of multi-source data fusion and analysis algorithms in real estate management and urban optimization. Specifically, we give priority to literature that explicitly mentions the application of multi-source data (such as sensor data, social media data, citizen feedback data, GIS data, etc.) in specific scenarios such as smart city real estate evaluation, market analysis, investment decision-making, urban planning, and resource allocation optimization. We exclude literature that only generally mentions the concept of data fusion or analysis but does not deeply involve the relevant application fields of smart cities. Studies that involve smart cities but are not closely related to real estate management and urban optimization are also excluded.

2.2.2. Research Value Judgment

We focus on selecting literature with high research value, which is mainly reflected in the following aspects. First, there are documents containing original research results, such as the proposal of new algorithms, innovative improvements to existing algorithms, and the verification of the effectiveness and feasibility of algorithms in actual smart city applications through empirical research. These documents can provide new ideas, methods, and technical support for our research and promote the theoretical and practical development of this field. Second, there are documents that can comprehensively and systematically review existing data fusion and analysis methods. Such documents not only provide detailed introductions to various algorithms but also deeply analyze their advantages and disadvantages, applicable scenarios, and problems faced in practical applications. These help us to fully understand the research status of this field, avoid repeated research, and provide references for subsequent research. For example, a review article that conducts a comprehensive comparative analysis of the application effects of multiple data fusion algorithms in smart city traffic management can help us quickly grasp the performance of different algorithms in this specific field, thereby providing a basis for us to choose the appropriate algorithm in our research.

2.2.3. Literature Quality Consideration

In terms of literature quality, we mainly use the source of literature and peer review as important judgment criteria. Articles published in peer-reviewed journals are preferred because these journals usually have a strict review process that can ensure the academic quality of the literature and the reliability of the research. During the peer review process, experts and scholars will conduct a comprehensive evaluation and review of the research methods, data analysis, and rationality of conclusions of the paper to ensure that the paper meets academic norms and research requirements. At the same time, we will also pay attention to papers from well-known academic conferences. Although the review process of conference papers may be slightly looser than that of journals, many high-level academic conferences have also attracted experts in many fields to participate in the review, and their papers can often reflect the latest research trends and cutting-edge achievements in the field. In addition, we will include high-quality research reports published by some research institutions or government departments in the research scope after comprehensive evaluation based on the rigor of their research, the reliability of data sources, and the rationality of their conclusions.

2.2.4. Timeliness Requirements

Considering the rapid development of technology in the field of smart city construction, data fusion, and analysis algorithms are constantly updated and iterated. In order to ensure that the research can reflect the latest progress and trends in this field, we limit the publication time of the literature to the past ten years. The literature within this period can better reflect the current application status, problems, and future development direction of multi-source data fusion and analysis algorithms in smart city construction at the current technical level. Although early literature may have certain reference value in terms of theoretical basis, due to the rapid development of technology, its research content and methods may no longer meet the needs of modern smart city construction. By focusing on literature in the past ten years, we can capture emerging technologies, innovative applications, and research hotspots in the field in a timely manner, providing more practical references for our research.

3. Multi-Source Data in the Construction of Smart City

3.1. Data Sources

Smart city construction involves a variety of multi-source data, mainly including sensor data, social media data, citizen feedback data, and GIS data [9]. These data provide rich information resources for real estate management and urban optimization. The comparison of these four types of data is shown in

Table 1. Comparison of sensor data, social media data, citizen feedback data, and GIS data.

Data Types	Data Source	Features	Application	Challenge
Sensor data	Sensors for weather, traffic flow, environmental monitoring, etc.	Real-time, dynamic, and accurate	Real estate evaluation and urban planning	Large amount of data, complex processing, privacy protection
Social media data	Data from social platforms such as Weibo, WeChat, and TikTok	Real-time, interactive, and diverse	Dynamic demand analysis, traffic management, public services	Authenticity is difficult to guarantee, privacy issues
Citizen feedback data	Public service platform, government microblog, mayor’s mailbox, etc.	Authenticity, timeliness, and diversity	Government decision-making, management effectiveness evaluation, public services	Information overload, difficulty in distinguishing true from false, and strong subjectivity
GIS Data	Surveying and mapping data, remote sensing images, and intelligent sensing equipment	Rich spatial and attribute data	Urban planning, real estate management	Data quality, updating, and sharing issues

.

Sensor data comes from various sensors, such as meteorology, traffic flow, and environmental monitoring. Sensors are strategically deployed across the city to monitor real-time information on the urban environment, traffic, and population distribution. These sensors collect data that is then transmitted to a centralized data-processing center or cloud infrastructure. At this hub, the data undergoes aggregation, analysis, and interpretation using advanced algorithms to extract meaningful insights. This centralized approach ensures efficient data management and allows for the integration of data from various sources, enabling comprehensive urban monitoring and decision-making. The processed information is subsequently used to inform urban planning, optimize resource allocation, and enhance the overall efficiency of city operations. Sensor data provides a scientific basis for evaluating real estate value, optimizing resource allocation, and improving urban management. These data can also realize real-time monitoring and early warning. For example, environmental monitoring sensors can detect air quality, water quality, etc., and issue timely warnings when abnormalities are found [10]. However, sensor data faces problems such as large data volume and obvious spatiotemporal characteristics, which require efficient processing and privacy protection measures [11].

Social media data includes user-generated content, social relationships, and geographic location information and is sourced from various social platforms [12]. These data reflect the dynamic needs of residents and can assist in decision-making in urban planning, traffic management, and public services. Social media data have challenges, such as difficulty in ensuring quality and authenticity, as well as privacy protection. Therefore, strict screening and verification are required during use, and technical means such as desensitization and encryption are used to protect privacy.

Citizen feedback data are collected through channels such as urban public service platforms, government microblogs, mayor mailboxes, and online questionnaires, reflecting residents’ needs and satisfaction with urban services. These data are authentic, reliable, and timely, covering a wide range of fields. Citizen feedback data needs to be cleaned, preprocessed, and analyzed in combination with other data sources to provide decision support, evaluate urban management effectiveness, and promote public service innovation. However, problems such as information overload and difficulty in distinguishing authenticity in citizen feedback data need to be further resolved [13].

GIS (Geographic Information System) data includes spatial data and non-spatial data, involving information such as the geographic location, shape, size, etc., of surface objects, as well as attribute information related to geographic location, such as population statistics, traffic flow, and land use [14]. These data can improve the scientific nature and decision support of urban planning and real estate management through accurate spatial information collection, processing, analysis, and display, but GIS data acquisition and processing face problems such as data quality and sharing and require the support of advanced data fusion and analysis algorithms [15].

3.2. Characteristics and Challenges of the Data

In the construction of smart cities, the spatiotemporal characteristics of multi-source data are its core attributes, which are of great significance to real estate management and urban optimization [16]. Spatiotemporal characteristics reflect the changing patterns of data in time and space dimensions. Through the analysis of time series data, the patterns of urban development can be mined and provide a decision-making basis for real estate management. At the same time, data in the time dimension need to fully consider real-time nature in order to grasp urban dynamics accurately [17]. In the spatial dimension, multi-source data show significant geographical distribution characteristics. Analysis of spatial data can reveal the laws of urban land use, traffic conditions, environmental pollution, etc., and provide a scientific basis for urban planning and optimization [18]. In addition, the coupling relationship between spatiotemporal characteristics is also an issue that cannot be ignored, requiring that the interconnection between time and space be fully considered when processing data.

At the same time, the fusion and analysis of multi-source data are the keys to improving management efficiency and decision-making quality in smart city construction, but data quality and consistency issues are major challenges in achieving the goal [19]. Data quality involves accuracy, completeness, reliability, timeliness, and availability, while data consistency is related to the unified standards and compatibility of data from different sources, formats, and types during the fusion process.

In addition, in the construction of smart cities, multi-source data fusion and analysis also face severe challenges in data privacy and security. With the increasing number of data sources, it is crucial to ensure that data security and personal privacy are not violated. In the data collection stage, unauthorized data collection may infringe on personal privacy. In the data storage and transmission stage, there are risks such as hacker attacks and data leakage. Sensitive information may also be accidentally leaked during the data analysis process [20].

4. Multi-Source Data Fusion Technology and Algorithm

4.1. Principles and Methods of Data Fusion Technology

Data fusion refers to the integration of data from different sources, formats, scales, dimensions, and properties to form a unified and coherent data set, thereby providing more comprehensive and accurate information support for real estate management and urban optimization [21]. The principle of data fusion technology is mainly to improve data quality and analysis accuracy by preprocessing and matching multi-source data [22]. Data preprocessing is the basis of multi-source data fusion, which aims to eliminate inconsistency, redundancy, and noise in the data and ensure the reliability of the analysis results. The process includes data cleaning, integration, conversion, and normalization to solve the problem of data inconsistency from different sources. For example, ensuring data integrity involves removing duplicate records and correcting erroneous data through a combination of automated checks, such as range validation and consistency checks, and manual verification by experts. This process efficiently identifies data errors by comparing the data against expected patterns, known values, and logical constraints. Data integration addresses the semantic heterogeneity of different data sources, such as sensors and GIS data, by standardizing data formats and creating a common vocabulary or schema. Data conversion then adjusts the original data to a format suitable for analysis, ensuring that the data are not only clean but also compatible with the analytical tools and methods used in our study. In addition, feature extraction is an important part of improving analysis efficiency, and key patterns are extracted by applying techniques such as principal component analysis. Data dimensionality reduction and outlier detection can further improve computing efficiency, ensure data quality, and provide a solid foundation for subsequent analysis [23].

In the process of data matching and association, the key is to effectively identify and link information from different sources to ensure the accuracy and completeness of the data. This link uses algorithms such as pattern recognition, similarity measurement, and entity resolution to establish a unified data view, eliminate data silos, and improve the effectiveness of decision support [24]. Data matching requires careful preprocessing, and then rule-based, similarity-based, or machine-learning methods can be used to address the challenges of data format heterogeneity, semantic differences, and large data size. With the development of technology, researchers have developed advanced algorithms such as distributed computing and graph computing for real-time dynamic data matching while also paying attention to the need to protect personal privacy during data fusion. In short, data fusion technology provides strong data support for real estate management and urban optimization by effectively integrating information from different data sources and promoting the sustainable development of smart cities.

In the construction of smart cities, multi-source data fusion technology is crucial, and data fusion algorithms are the key means to achieve this goal. These algorithms effectively integrate, complement, and optimize multi-source data to improve their availability, accuracy, and reliability [25]. The main types of data fusion algorithms include methods based on probability statistics, evidence theory, fuzzy set theory, neural networks, and machine learning. Each method has its own characteristics and is suitable for different scenarios and data types [26]. The types and comparisons of data fusion algorithms are shown in

Table 2. Types and comparison of data fusion algorithms.

Algorithm Type	Main Features	Applicable Scenarios
Fusion algorithm based on probability statistics [27]	By calculating the probability distribution of data sources, weighted fusion is achieved, which is suitable for processing uncertain and random data.	Fusion of environmental data and citizen feedback in real estate management and urban optimization.
An integrated fusion algorithm based on evidence theory [28]	By constructing a body of evidence, information from different data sources is synthesized to handle contradictory and conflicting data.	When faced with conflicting data from multiple sources, such as reducing the impact of erroneous information in decision making.
Fusion algorithm based on fuzzy set theory [29]	Deal with uncertainty and fuzzy data to improve the interpretability and practicality of the results.	Process feedback data with a high degree of neutrality, such as citizen opinions.
Neural network-based algorithms [30]	Automatically learn data features with adaptability and generalization capabilities, suitable for large-scale, high-dimensional data.	Real estate management combines sensors and GIS data to support urban optimization.
Algorithms based on machine learning [31]	Data fusion is achieved through model training, which is adapted to specific goals and is powerful in processing complex data relationships.	Extract useful information and analyze complex relationships to improve real estate management and urban optimization.
Distributed Consensus Gossip Algorithms [32]	Including Randomized gossip algorithm (RG), Geographic gossip algorithm (GG), Broadcast gossip algorithm (BG), Push-Pull protocol (PP) and Push-Sum protocol (PS). Applicable to network size estimation in multi-agent systems. Determine or estimate various aggregation functions in an iterative manner.	Real-time traffic monitoring, energy management, environmental monitoring, public safety, real estate valuation, urban planning, and energy optimization of smart buildings to improve urban management efficiency and residents’ quality of life.
Industrial Internet of Things, IIoT [33]	The data are analyzed by analyzing the Number of Pixels per Change Rate (NPCR) and entropy information. Public key encryption is performed using the Edwards-curve Digital Signature Algorithm (EdDSA) and a Schnorr signature variant based on the Edwards curve.	Real estate image encryption in IIoT environments. Check the confidentiality and security of real estate data by storing encrypted pixel values in the blockchain.

.

These data fusion algorithms provide strong support for the construction of smart cities. Through reasonable selection and application, the application value of data can be significantly improved [34]. However, they still face challenges in implementation, such as data quality, consistency, and real-time performance.

4.2. Application of Multi-Source Data Fusion Algorithm in Real Estate Management

In the construction of smart cities for real estate management, the application of multi-source data fusion algorithms is crucial. Among them, the fusion of sensor data and GIS data provides real-time environmental monitoring and urban spatial analysis. Integrating real-time monitoring information and geospatial data promotes application scenarios such as urban planning, disaster warning, and ecological monitoring and improves decision-making efficiency [35]. However, the fusion process faces challenges such as data quality and privacy security. At the same time, the combination of social media data and citizen feedback data enables managers to have a more comprehensive understanding of the use of urban real estate and make scientific decisions. Through data preprocessing and fusion algorithms, the real estate market can be analyzed, urban facilities can be optimized, and community governance and public services can be improved. These two data fusion schemes provide strong support for real estate management, but in practical applications, attention should still be paid to data quality and compliance to achieve the sustainable development of smart cities. The comparison of application cases of sensor data and GIS data, social media data, and citizen feedback data in real estate management is shown in

Table 3. Comparison of application cases of sensor data and GIS data, social media data, and citizen feedback data in real estate management.

Data Fusion Type	Application Cases
Fusion of sensor data and GIS data [36]	1. Real-time monitoring of urban area environment to support decision-making and urban planning.
	2. Extract and analyze geographic features, such as land use and traffic conditions, to assess real estate value.
	3. Optimize traffic management, adjust traffic signals through data fusion, and improve circulation efficiency.
Fusion of social media data and citizen feedback data [37]	1. Real estate market analysis, combined with housing supply and residents’ satisfaction, provides a basis for policy-making.
	2. Optimize urban supporting facilities and analyze the service and transportation convenience of different areas.
	3. Community governance: timely addressing of environmental and safety issues and improving the quality of life of residents.
	4. Optimize public services, analyze citizen feedback, and improve the pertinence and effectiveness of government services.

.

In the construction of smart cities, multi-source data fusion algorithms are crucial for real estate management and urban optimization but face many challenges in actual operations [38]. For example, heterogeneous data sources (such as sensors, social media, citizen feedback, and GIS data) have significant differences in format, scale, semantics, and quality, which hinders effective integration. To this end, data preprocessing techniques (such as cleaning and standardization) need to be used to improve data quality and consistency and lay a good foundation for subsequent analysis. When processing massive data, the computational complexity and storage requirements are high, which may lead to performance bottlenecks. Therefore, distributed and cloud computing technologies can be used to achieve parallel processing and improve efficiency while reducing storage requirements through data compression and indexing [39]. In the context of smart city operations, the protection of privacy and data security is paramount [1]. We must be vigilant against topical security threats such as cyber-attacks, data breaches, and unauthorized access, which can compromise sensitive information and disrupt urban services [2]. These threats are dangerous as they can lead to financial losses, legal repercussions, and a loss of public trust in smart city initiatives. To mitigate these risks, it is essential to adopt robust encryption techniques and stringent access control strategies to safeguard data integrity and confidentiality [20]. Compliance with data protection laws and regulations, such as the General Data Protection Regulation (GDPR), is also crucial to ensure that data handling practices are transparent and secure. Furthermore, to address the dynamic nature and real-time requirements of smart city data, we propose the introduction of advanced spatio-temporal data analysis methods. These methods require inputs such as geospatial coordinates, temporal data stamps, and attribute information related to urban features, which enable the algorithms to analyze and model the spatial relationships and temporal dynamics within the city’s data. By integrating these inputs, the algorithms can provide insights into urban patterns and trends, supporting decision-making in areas like urban planning, traffic management, and environmental monitoring. Stream data-processing technology can be leveraged for rapid analysis, ensuring that decision-making is informed by the most current data available. For the scalability and adaptability of our algorithms, we advocate for the design of a modular fusion framework [26]. This framework, when integrated with machine-learning technology, can automatically optimize the data fusion process based on the characteristics of the data and specific business needs [28]. This approach not only enhances the efficiency of data processing but also strengthens the resilience of smart city systems against evolving security threats. These technical means will effectively improve data fusion performance and provide strong support for real estate management and urban optimization of smart cities, ensuring that the spatial statistical analysis algorithms are fed with the necessary inputs to deliver actionable insights.

5. Application of Data Analysis Algorithm in Urban Real Estate Management

5.1. Spatial Data Analysis Algorithm

Spatial statistical analysis algorithms are indispensable tools in the construction of smart cities. They mainly provide scientific and accurate decision support for urban real estate management and optimization through in-depth analysis of geospatial data [40]. This type of algorithm integrates knowledge from multiple fields, such as statistics, geography, and computer science, and uses the geographic information system (GIS) as a platform to effectively mine and analyze spatial data. Spatial statistical analysis algorithms include classic spatial autocorrelation analysis, spatial variogram analysis, and geostatistical analysis methods. These methods play a key role in revealing the spatial distribution characteristics of real estate, quantifying spatial correlation, and predicting spatial trends [41]. First, spatial autocorrelation analysis can detect the correlation between spatial data to determine whether a certain attribute value has a clustering characteristic in space. This helps to understand the spatial distribution laws of urban real estate value, type, and usage, thereby providing a scientific basis for the government to formulate land policies and plan urban layout. Secondly, spatial variogram analysis can describe the degree of variation and structural characteristics of spatial data within a certain range, which is of great significance for understanding the regional differences in the real estate market and formulating targeted regulatory measures. In addition, geostatistical analysis methods can interpolate and predict spatial data such as real estate prices and supply and demand conditions, providing a reference for market evaluation and investment decisions [42]. In addition, spatial statistical analysis algorithms can effectively deal with data quality and consistency issues when processing real estate data in the big data era [43]. By reasonably selecting or improving algorithms, the impact of outliers, missing values, etc., on analysis results can be reduced, and the accuracy of real estate management decisions can be improved. At the same time, spatial statistical analysis algorithms also have certain advantages in protecting data privacy and security and can provide reliable data support for urban managers.

As shown in

Table 4. Application of spatial statistical analysis algorithms in urban real estate management.

Algorithm Name	Algorithm Description	Application Scenario	Main Function
Kriging interpolation [44]	A spatial interpolation method based on variogram, used to predict spatially distributed data.	Land Price Forecast	Predict the land value in different regions and provide a basis for real estate valuation.
Spatial autocorrelation analysis [45]	Detect the spatial relationships between objects in a dataset to determine if there are clustering patterns.	Real estate market analysis	Identify hot areas in the real estate market.
Nearest neighbor analysis [46]	Calculate the distance between each point and its nearest point and analyze the uniformity of the point distribution.	Commercial network layout	Optimize the spatial layout of commercial outlets and improve market coverage efficiency.
Spatial clustering analysis [47]	The spatial data are grouped wherein the objects within the same group have a high similarity, while the objects between different groups have a low similarity.	Urban Plan	Identify areas of a city with similar characteristics to guide urban planning.
Buffer zone analysis [48]	Create buffers around specific geographic features to analyze the impact those features have on surrounding areas.	Environmental Monitoring	Assess the scope of impact of real estate development on the surrounding environment.
Road network analysis [49]	Analysis based on road network data is used to determine issues such as paths, distances, and accessibility.	Traffic Optimization	Optimize traffic flow and improve real estate accessibility.

, in practical applications, spatial statistical analysis algorithms have been successfully applied to many fields, such as urban planning, land consolidation, and real estate market monitoring. Through in-depth mining of urban real estate spatial data, these algorithms help to reveal the inherent laws of urban development and provide strong support for urban optimization. Spatial statistical analysis algorithms are of great value in the construction of smart cities and are expected to promote real estate management and urban optimization to a higher level.

At the same time, geospatial data mining algorithms, as a key technology, also play a vital role in the construction of smart cities. They mainly use spatial data in geographic information systems (GIS) to reveal the hidden laws and patterns in the data through mining and analysis, providing a scientific basis for urban real estate management and urban optimization. Geospatial data mining algorithms include many methods, such as spatial clustering, spatial association rule mining, spatial prediction, and spatial classification [50]. These algorithms effectively combine the geographical location, attribute characteristics, and time dimension of spatial data to provide intelligent support for the planning, evaluation, transaction, and supervision of urban real estate. The introduction and comparison of various geospatial data mining algorithms are shown in

Table 5. Introduction and comparison of various geospatial data mining algorithms.

Algorithm Name	Describe	Applicable Scenarios	Advantage	Shortcoming
Density-based spatial clustering algorithm (DBSCAN) [51]	Density-based clustering algorithm, which does not require the number of clusters to be specified in advance	Urban area division and hot spot analysis	Can identify clusters of arbitrary shapes	Sensitive to noise and outliers
K-means clustering algorithm [52]	Divide the spatial data points into K clusters wherein the distance between points in the cluster is minimized	Residential area planning, commercial area layout	The algorithm is simple, and the computational efficiency is high	The number of clusters needs to be specified in advance and is sensitive to noise and outliers
Grid-based spatial clustering algorithm [53]	Divide the space into grid cells, count the number of points in each cell, and form clusters	Urban traffic planning, crime data analysis	High efficiency in processing large data sets	The choice of grid size has a great influence on the results
Geospatial association rule mining algorithm [54]	Discover association rules in spatial data, such as neighbor relationships and proximity relationships	Urban planning, environmental monitoring	Potential spatial relationships can be discovered	High computational complexity, requiring a lot of computing resources

.

5.2. The Application of Deep-Learning Algorithm in Real Estate Data

As one of the core technologies in the field of artificial intelligence, the deep-learning algorithm is a method of simulating the structure and function of the human brain through multi-layer artificial neural networks, aiming to analyze, understand, and generate complex data [55]. As a deep-learning algorithm, convolutional neural network (CNN) has achieved remarkable results in the field of image processing, and its application in real estate image analysis has gradually shown its unique advantages [56]. Real estate image analysis mainly includes real estate image recognition, land use classification, building damage assessment, etc. CNN provides powerful technical support for real estate management and urban optimization by extracting features and classifying these images. First of all, in terms of real estate image recognition, CNN can automatically extract features in real estate images, such as architectural style, floor height, surrounding environment, etc., to achieve accurate identification of property types. The application of this technology can help improve the efficiency of real estate registration, evaluation, and transactions and provide more accurate data support for the real estate market. Secondly, land use classification is an important link in urban planning and land resource management. Using CNN to classify remote sensing images can quickly and accurately identify various land use types, such as residential, commercial, industrial, greening, etc. [57]. This helps government departments formulate reasonable land use policies and optimize urban spatial layout. In addition, CNN also has broad application prospects in building damage assessment. By analyzing building images after natural disasters such as earthquakes and floods, CNN can identify the degree of damage to buildings and provide strong support for post-disaster rescue and reconstruction. The application of this technology can help improve rescue efficiency and reduce disaster losses.

In the process of real estate image analysis, the core advantages of CNN are reflected in the following aspects:

• Feature extraction capability: CNN can automatically extract local features in the image through the combination of convolutional layers and pooling layers and gradually abstract higher-level features, thereby improving the accuracy and robustness of image analysis.
• Parameter sharing: The weight-sharing strategy in CNN greatly reduces the number of model parameters, reduces the computational complexity, and improves the operation speed.
• Applicable to complex scenarios: CNN can process complex image data, such as deformation, rotation, scaling, etc., which gives it a wide range of potential applications in real estate image analysis.
• End-to-end learning: CNN adopts an end-to-end learning method, which avoids the cumbersome features of engineering in traditional image analysis and improves the analysis efficiency.

As shown in

Table 6. Example of using convolutional neural networks in real estate image analysis.

Method	Describe	Application Scenario	Effect
CNN classification [58]	Use CNN to classify real estate images into types such as residential, commercial, and industrial.	Automatic identification of real estate types	High accuracy, easy to apply on a large scale
CNN feature extraction [56]	Extract real estate image features for subsequent image recognition or regression analysis.	Value Assessment Assistance	Extracting features is effective and helps evaluate the model
Semantic segmentation [59]	Classify each pixel in the image to distinguish between different structures or objects in the property.	Building Structural Analysis	High-resolution image analysis, down to the specific structure
Target detection [60]	Detect and locate specific targets in images, such as vehicles, pedestrians, etc.	Security Monitoring	Enhance real estate safety management and detect abnormalities in a timely manner

, the application of convolutional neural network (CNN) in real estate image analysis has significant advantages, providing strong support for real estate management and urban optimization [30]. With the continuous advancement of deep-learning technology, CNN will be more widely used in the field of real estate image analysis in the future, making greater contributions to the construction of smart cities. To ensure the privacy and security of citizens within these applications, we have integrated stringent data protection measures [29]. These include the use of anonymization techniques to preprocess images, ensuring that no personal ID information is extracted or stored [31]. We also focus on analyzing public spaces and aggregated data, which do not compromise individual privacy. Furthermore, the deployment of these algorithms is subject to regulatory oversight to prevent misuse and to uphold ethical standards in data processing. By adopting these practices, we can leverage the power of CNN for urban analysis while maintaining the highest standards of privacy protection.

6. Synergism Between Algorithms and Smart City Optimization

6.1. Collaborative Mechanism of Multi-Source Data Fusion and Data Analysis Algorithms

The principle and implementation approach of algorithm collaboration aims to improve the efficiency and accuracy of urban management and optimization by combining different types of data fusion technologies and data analysis methods [61]. Specifically, algorithm collaboration is not a simple combination but a mechanism of mutual dependence and enhancement, such that each algorithm can complement the other in terms of function. Data fusion technology eliminates information silos and provides a comprehensive and consistent data foundation for analysis, while data analysis algorithms mine potential patterns and trends, thereby providing decision support for urban real estate management [62]. Its implementation approach mainly includes building a multi-level data fusion architecture, ensuring data quality and consistency through data preprocessing, using a variety of data analysis methods, combining traditional statistics with advanced deep-learning technology to mine valuable information, promoting interdisciplinary cooperation, strengthening the integration of algorithms in different fields, and dynamically adjusting and optimizing algorithms to adapt to changing urban needs and ensure real-time response and stability. Through these methods, algorithm collaboration not only improves the performance of a single algorithm but also provides a scientific and refined decision-making basis for the sustainable development of smart cities [63].

6.2. The Application Effect of Algorithm Collaboration in Real Estate Management in Smart City

In the construction of smart cities, real estate management and urban optimization rely on the fusion and analysis of multi-source data [64]. Improving the accuracy and efficiency of data processing has become a key link, which is of great significance for exploring the deep value of data and realizing efficient urban management. To achieve this goal, we need to start with the following aspects. First, we optimize the data preprocessing process to ensure data quality, which includes accuracy, completeness, and consistency. Data quality is compromised by factors such as sensor errors due to environmental conditions. By addressing these issues through cleaning, denoising, and outlier detection, we provide a reliable foundation for subsequent analysis. Secondly, efficient data fusion technology, such as the data fusion method based on spatiotemporal association rules, can be adopted to effectively integrate data from different sources, formats, and scales to improve data availability. In addition, introducing advanced data analysis algorithms, such as deep-learning and machine-learning algorithms, to intelligently mine the fused data and discover potential laws and trends provides strong support for real estate management and urban optimization. At the same time, attention should be paid to the synergy between algorithms to achieve complementary advantages between data fusion and analysis algorithms. For example, combining spatial statistical analysis with geospatial data mining can improve the accuracy of the analysis of real estate spatial distribution characteristics, and deep-learning algorithms can further improve data processing efficiency when processing complex data relationships. In addition, by building an adaptive algorithm adjustment mechanism, the algorithm can automatically optimize parameters according to data characteristics, thereby improving the overall data processing performance. In order to further improve the accuracy and efficiency of data processing, the following points should be paid attention to: first, establish a complete data quality control system to ensure the accuracy of data throughout the entire processing process; second, strengthen interdisciplinary research, draw on technical achievements in related fields, and provide new ideas for smart city data fusion and analysis; third, pay attention to the real-time and dynamic adaptability of algorithms to cope with the ever-changing urban environment and real estate market. Improving the accuracy and efficiency of data processing is the core task of multi-source data fusion and analysis in smart city construction.

Of course, in the real estate management of smart cities, the synergy of multi-source data fusion and analysis algorithms is of great significance for promoting the rational allocation of urban resources [65]. By integrating various sensor data, social media information, citizen feedback, and geographic information system data, combined with advanced data fusion technology and analysis algorithms, we can deeply explore the spatiotemporal characteristics, utilization status, and potential value of urban real estate. This fusion not only improves the availability and accuracy of the data but also provides comprehensive decision-making support for urban planners and helps to efficiently allocate urban resources. First, the fusion of multi-source data can fully reveal the spatial distribution characteristics and utilization efficiency of urban real estate, providing a scientific basis for urban planning. Through accurate analysis of the current status of real estate, we can identify idle resources, over-utilized areas, and development potential areas in the city, thereby guiding government departments to reasonably adjust the nature of land use and optimize spatial layout. In addition, combined with spatiotemporal data analysis, urban development trends can be predicted, providing scientific guidance for future infrastructure construction and public service facility layout and avoiding resource waste [66]. Secondly, the application of data analysis algorithms helps to explore the potential demand of the urban real estate market and provides a basis for market regulation. By analyzing real estate transaction data, rental prices, and other information, we can understand the market supply and demand situation, provide support for the government in formulating differentiated regulatory policies, and encourage resources to flow to more efficient areas. At the same time, this also helps to guide the rational investment of social capital and stimulate market vitality. Furthermore, the synergy of multi-source data fusion and analysis algorithms in real estate management helps to improve the operational efficiency of urban infrastructure. For example, through intelligent analysis algorithms, traffic flow, energy consumption, and other data can be monitored and analyzed in real time, providing strong support for infrastructure planning, construction, and management and achieving optimal resource allocation. In addition, algorithm synergy is also of great significance in optimizing the urban ecological environment [67]. Combined with environmental monitoring data, land use data, etc., the quality of the urban ecological environment can be evaluated, and a decision-making basis can be provided for environmental protection and governance. By rationally adjusting land use methods and optimizing ecological space layout, it is helpful to improve the quality of the urban environment and achieve sustainable development.

7. Future Prospects of Multi-Source Data Fusion and Analysis Algorithms in Smart Cities

7.1. Data Quality and Real-Time Challenges That Need to Be Overcome

With the accelerated advancement of smart city construction, multi-source data fusion and analysis algorithms have increasingly stringent requirements for data quality and real-time performance. At present, uneven data quality and insufficient real-time processing capabilities have become key factors restricting the effectiveness of algorithms. In terms of data quality, multi-source data often have problems such as noise interference, data missing, and data inconsistency, which seriously affect the accuracy of fusion and analysis results. For example, sensor data may produce outliers due to equipment failure or environmental factors; social media data may contain a large amount of false information or irrelevant content; citizen feedback data are greatly affected by subjective factors; GIS data may have geographic information deviations caused by untimely updates. In terms of real-time performance, despite the continuous advancement of technology, it is still difficult for algorithms to process and respond immediately when faced with massive and high-speed data, and they cannot meet the urgent needs of smart cities for real-time decision-making, such as real-time diversion of traffic congestion and emergency response to emergencies.

In the future, in order to improve data quality, a more complete data cleaning, verification, and updating mechanism needs to be established. On the one hand, intelligent data-cleaning algorithms can be developed to automatically identify and correct errors and outliers in the data. Moreover, machine-learning technology can be used to perform semantic understanding and consistency verification of data to ensure the compatibility and accuracy of data from different sources. On the other hand, establishing real-time data collection channels and dynamic update mechanisms can strengthen the timeliness of data updates and ensure that data can always reflect the latest status of the city. In terms of improving real-time performance, efficient data-processing architectures and algorithm optimization strategies, such as using distributed computing, stream computing, and other technologies, to achieve parallel processing and real-time analysis of data ensure that the algorithm can quickly respond to dynamic changes in the city and provide timely and accurate data support for urban management decisions [68].

7.2. Exploration Direction of Deep Mining and Association Analysis

At present, there is still much room for improvement in multi-source data fusion algorithms in mining the potential value of data. There are rich intrinsic connections between different data sources, but most of the current analysis methods only stay on the surface and fail to fully reveal their deep-level connections and laws. For example, there may be some spatiotemporal correlation between sensor data and social media data, reflecting the real-time response of urban residents to environmental changes, but existing algorithms are difficult to effectively capture and use this correlation for in-depth analysis.

Future research should focus on developing more powerful deep mining and correlation analysis technologies [69]. By integrating multidisciplinary knowledge, such as data mining, machine learning, and statistics, complex data models can be constructed to explore hidden patterns and potential relationships between different data types. For example, deep-learning algorithms can be used to jointly analyze massive amounts of sensor data and social media text data to mine the intrinsic connection between changes in residents’ emotions and fluctuations in environmental indicators, providing a more forward-looking decision-making basis for urban management. At the same time, attention should be paid to the visualization of correlation analysis results, presenting complex relationships to urban managers in an intuitive and easy-to-understand way—such that they can better understand the urban operation mechanism and formulate more scientific and reasonable policies and plans.

7.3. The Coordinated Development of Interdisciplinary Research and Privacy Protection

The application of multi-source data fusion and analysis algorithms in smart city construction involves multiple disciplines, but interdisciplinary research is not yet in-depth, and the synergy between disciplines has not been fully utilized. There is a disconnect between urban planning, sociology, computer science, and other disciplines in data understanding, application needs, and technical implementation, which makes it difficult to perfectly match algorithm design with actual urban management needs. At the same time, with the increasing popularity of data collection and use, privacy protection and data security issues have become increasingly prominent, becoming an important factor restricting the promotion and application of algorithms.

In the future, interdisciplinary research will become a key driving force for the development of this field. Experts and scholars from different disciplines are encouraged to work closely to conduct research projects and break down disciplinary barriers. Urban planners and data analysts should work together to optimize algorithm design based on urban development goals and spatial layout needs to better serve urban planning and resource allocation. Sociologists should participate in algorithm evaluation to ensure that algorithm applications positively impact society from the perspectives of social fairness and resident satisfaction. In terms of privacy protection, laws, regulations, boundaries, and responsibilities of data use must be strengthened and clarified. At the same time, we should develop innovative privacy protection technologies, such as encryption algorithms, anonymization processing technologies, and data desensitization technologies, to ensure that personal privacy is not violated during the data fusion and analysis process and to achieve a balance between data utilization and privacy protection [70].

7.4. Innovation Path of Algorithm Collaboration and Application Scenario Expansion

At present, although data fusion and analysis algorithms have made certain progress in their respective fields, there are still deficiencies in algorithm collaboration and application scenario expansion. There is a lack of effective integration mechanisms between different algorithms, which makes it impossible to give full play to their respective advantages and achieve synergy and efficiency. In addition, the existing application scenarios are mainly concentrated in limited fields such as real estate management, transportation, and environment and fail to fully cover all aspects of smart city construction.

In the future, efforts should be made to build a multi-algorithm collaborative framework to achieve seamless docking and collaborative work between different algorithms by optimizing algorithm combinations and parameter configurations [71]. For example, to jointly improve the accuracy of real estate value assessment, spatial statistical analysis algorithms can be combined with deep-learning algorithms—spatial statistical analysis algorithms can be used to process the structural characteristics of geographic spatial data and deep-learning algorithms can be used to mine data nonlinear relationships. At the same time, algorithm application scenarios should be actively expanded. For example, in the public service field, multi-source data fusion and analysis algorithms can be used to optimize the allocation of educational resources, improve the efficiency of medical services, and improve the quality of community elderly care services; in social management, they can help with social security prevention and control, urban emergency command, public opinion monitoring and guidance, etc.; in the cultural tourism industry, tourist route optimization, cultural heritage protection and development can be achieved by analyzing tourist behavior data and the distribution of urban cultural resources. By continuously expanding application scenarios, we can give full play to the all-round support role of multi-source data fusion and analysis algorithms in the construction of smart cities, improve the quality of life and happiness of urban residents, and promote sustainable urban development [72].

8. Conclusions

With the acceleration of urbanization, the traditional real estate management model faces many challenges, including information asymmetry, inefficient resource utilization, and increased environmental pressure. Our study systematically and comprehensively reviews the multi-source data fusion and analysis algorithms in smart city construction, combines the characteristics and application scenarios of different data sources, and deeply explores its key role in real estate management and urban optimization. This study not only provides theoretical support for scholars in related fields but also provides practical guidance for urban managers, which has important practical significance.

Through a review of existing literature, we found that multi-source data fusion and analysis algorithms not only enhance the scientific nature of urban planning but also improve the efficiency of real estate management. Specifically, data fusion technology effectively eliminates information islands, enables various types of data to be seamlessly integrated, and forms a consistent data foundation, while data analysis algorithms reveal the laws and trends of urban development by mining the potential correlations between data, providing strong support for policy making. In addition, the synergy mechanism between algorithms enhances the performance of a single algorithm—by integrating the advantages of different technologies—and promotes the sustainable development of smart cities. Therefore, the development potential of multi-source data fusion and analysis algorithms in smart city construction is huge. The application of its concepts and technologies will effectively promote the continuous progress of real estate management and urban optimization and help achieve efficient, intelligent, and sustainable urban development goals.