1.1 Realty management

Realty or property management is a process that supervises an organization’s or industry’s properties. Realty management is an important task that manages properties based on priorities and commitments. The systems manage legal disputes and interests of the properties [1]. Realty Management provides various amenities to the users that provide effective services. The property management system maintains, controls, oversees and processes real estate or physical properties. The market value of the properties is also maintained by management systems [2]. Various techniques and technologies are used for realty management systems. Cloud-based techniques are most widely used for property management systems. Cloud computing identifies the flexible solutions required to maintain the properties [3, 4]. A cloud computing system detects the economic scale of a property and provides necessary information for management systems. Machine learning (ML) based methods are used for property management systems. The support vector machine (SVM) algorithm is usually used as an ML technique that improves accuracy in real-time management systems. SVM predicts the functions and operations required to provide better performance in the management process [5]. Location, economy, government policy, and consumer preferences may impact real estate markets. Adding domain knowledge and context-specific elements to the learning process is paramount.

Cloud computing is a significant component of real estate management since it facilitates providing computer services via the internet. This technology enables data storage, processing, and analysis on remote servers. This technology allows for the retrieval and examination of large amounts of organized and unorganized data from many sources, such as property listings, market trends, and consumer behaviour. Artificial intelligence (AI) can be used in the real estate industry to automate property valuations, optimize pricing tactics, and uncover investment opportunities. This is achieved using algorithms that can learn from data and generate predictions or suggestions. Machine learning algorithms, which are a subset of artificial intelligence (AI), demonstrate exceptional proficiency in identifying patterns and may be trained using past data to make predictions about future trends in the real estate market, anticipate property prices, and provide recommendations for suitable renovations or improvements to optimize returns. An example of this would be a machine learning model trained using historical sales data. This model would be capable of analyzing various attributes of a property and its surrounding area to estimate its prospective market worth. This would greatly assist real estate agents and investors in making well-informed judgments.

1.2 Importance of big data in realty management

Big data extraction is a process that extracts data from other systems or sources. Big data extraction identifies the important data and variables required for further application processes. The big data extraction method provides feasible information to execute a certain system task. Big data extraction is used for realty management systems. Big data is mainly used to decrease the complexity and latency in the data identification progression [6]. Big data analytic techniques identify the optimal information from a huge amount of data. Big data extraction first analyzes the database containing relevant data for property management systems [7]. Big data extraction extracts important features from other systems, improving the systems’ flexibility and reliability range. Users widely use Internet of Things (IoT) based property management systems. IoT uses various methods that predict the contents for further processes. Commercial real estate (CRE) contains important operational functions. Big data extracts data from CRE that reduces the complexity of realty management systems [8, 9]. The efficacy and acceptance of the big data and AI decision-making system are directly affected by how user-friendly the interface is for realty managers to engage with it. The ease of use, data entry, result interpretation, and decision-making for realty managers is guaranteed by an intuitive interface. Realty management decision-making is revolutionized by integrating big data and artificial intelligence (AI) technology. These improved capabilities boost efficiency, accuracy, and strategic insights.

The quality, completeness, and diversity of real estate data might vary. The accuracy and reliability of the insights obtained by artificial intelligence models might be compromised by using inaccurate or inadequate data. A decision-making framework incorporating big data and AI in realty management aims to leverage large volumes of diverse data to inform decision-making processes through advanced analytics and machine learning. Improving realty management capabilities with big data integration entails making better decisions, running more efficient operations, and managing assets more effectively by harnessing massive amounts of varied, complicated, and real-time data. Evaluating and measuring realty management abilities often entails looking at several aspects or criteria to see how well real estate management operations work. The ability to competently and efficiently manage real estate assets, properties, and transactions is known as realty management ability. Maximizing profits and ensuring the optimum use of real estate assets include a variety of operations linked to the purchase, development, operation, and disposal of real estate.

1.3 Importance of artificial intelligence in realty management

Artificial intelligence (AI) is widely utilized in different fields and applications. AI technique is primarily utilized to enhance the performance and efficiency of the systems. AI is used in the decision-making process of realty management systems [10]. Decision-making is a complicated task to execute in the realty management system. The decision-making process requires accurate datasets to make the proper decision [11]. AI-based decision-making algorithm is used in realty management. Important key values and characteristics are identified by AI using the feature extraction technique [12]. Feature extraction extracts the variables for decision-making, decreasing the computation process’s latency. The AI-based algorithm improves overall decision-making accuracy, enhancing the effectiveness of property management systems [13]. Artificial neural network (ANN) is utilized for the decision-making process. ANN analyzes the data that are required for decision-making. ANN identifies the key features from the database. ANN decreases the latency and complexity in the computation procedure, enhancing the decision-making process’s feasibility range [14, 15]. Big data handling and processing with precision for reliable property management through diverse considerations requires complex computations. With the help of big data and AI, property managers can sort through data peaks and make informed decisions. Advanced analytics and machine learning algorithms may uncover patterns, trends, and insights that are not obvious through conventional approaches. Implementing stringent restrictions on access following the concept of least privilege. Verify user identities and allocate access rights following task responsibilities. Remain at the forefront of the company’s ever-evolving demands by regularly reviewing and updating access privileges. A decision-making framework incorporating big data and AI in realty management aims to leverage large volumes of diverse data to inform decision-making processes through advanced analytics and machine learning. Real estate market trends may be foretold with AI algorithms by analyzing past data. The accuracy of property evaluations is ensured by AI algorithms considering many parameters, such as property attributes, market circumstances, and previous sales data. Cleansing, validating, and enriching data are all part of strong data quality assurance procedures. To guarantee high-quality datasets for analysis, this study will use AI techniques to manage incomplete or noisy data.

The issue of enhancing real estate management proficiency emerges within real estate management, wherein proficient management of property-related data is of paramount importance. Multiple parties, such as property owners, real estate agents, property managers, and tenants, are impacted by this matter. Efforts have been undertaken to address this issue by advancing diverse data processing methodologies and technology. One method is implementing knowledge-dependent data processing schemes (KDPS) to improve real estate data management by utilizing domain-specific knowledge and norms. In the processing and analysis of property-related data, these schemes employ many techniques, including rule-based systems, expert systems, and ontology-based approaches, to collect and apply domain knowledge effectively. Furthermore, researchers have also investigated advancements in fields such as machine learning and artificial intelligence to enhance the precision and effectiveness of real estate data processing.

1.4 Contribution of this study

Designing and implementing a Knowledge-Dependent Data Processing Scheme (KDPS) for realty management requires careful attention to accessibility, privacy, security, ethical implications, and other concerns. The performance metrics include stagnancy detection, success rate, training ratio, and reduced processing time. Several key performance indicators (KPIs) and metrics may be used to measure and evaluate the progress in realty management capabilities, especially with installing a Knowledge-Dependent Data Processing Scheme (KDPS). The primary objective of the knowledge-dependent data processing scheme (KDPS) is to optimize real estate management by utilising domain-specific knowledge and data analysis methodologies. Data acquisition and organization encompass collecting and arranging pertinent information from diverse sources, including property listings, market trends, and client preferences. The data is further subjected to processing and analysis utilizing knowledge-based rules and algorithms that integrate domain experience derived from specialists in the real estate industry. The insights that have been retrieved are utilized to construct prediction models and decision support systems. These systems are designed to aid in property assessment, investment analysis, and client targeting tasks. By integrating data-driven insights with human experience, KDPS empowers real estate managers to make well-informed decisions, resulting in greater operational efficiency, optimized resource allocation, and improved customer satisfaction. The assessment should consider qualitative and quantitative metrics to provide a full picture of the effect. Real estate managers can foresee changes in the market because of data on market trends and economic factors such as inflation and interest rates. The value of a property may be better understood by looking at its sales history.

Considering its significance, this article introduces a knowledge-based processing scheme that is split and briefed as given below:

1. Introducing a knowledge-based processing scheme for validating property data upon multiple criteria such as market value, maintenance, and sales
2. Incorporating knowledge learning for improving the operations of realty systems towards flawless data handling and precise recommendations
3. Performing an extensive analysis using appropriate external data to verify its adaptability for the proposed scheme

The rest of the paper is organized as follows. Section 2 describes the various researchers opinions regarding the realty systems. Section 3 discusses the working process of the Knowledge-Dependent Data Processing scheme-based realty management process and the system’s efficiency evaluated in section 4. Conclusion described in Section 5.

2.1 Deep learning-based models for realty management

Wang et al. [16] developed a DL model for house price prediction utilizing heterogeneous information analysis. The proposed model is a joint self-attention model that extracts necessary data from the database. The price prediction process evaluates the availability of amenities, rapid transit, location, and property environment. The proposed model identifies the important datasets that are required for prediction. The suggested model attains high accuracy in prediction, improving the systems’ performance. However, the system requires additional effort to analyse a large data volume.

Chen et al. [17] introduced a semiconductor intellectual property (SIP) business model for value creation analysis. The key objective of the introduced model is to evaluate the actual value creation mechanism for SIP firms. The firms’ intellectual property (IP) is identified, producing optimal data for business models. The introduced model classifies the IP based on specific characteristics and priorities. The SIP business model is mostly used for owners’ entrepreneurship. The study offers insights into semiconductor intellectual property’s economic models and dynamic capacities. However, its generalizability may be limited due to its special focus on the semiconductor industry. The generalizability of the findings and consequences to industries subject to distinct dynamics, technical landscapes, and intellectual property considerations may be limited.

Su et al. [18] proposed building information modelling (BIM) and machine learning (ML) integration framework automated property evaluation. An information extraction method is used here to identify important information from the database. ML mainly detects the key values required for the evaluation process. The suggested framework enhances the information exchange process among individuals. Numerical outcomes demonstrate that the proposed framework decreases the error ratio in the evaluation process. The potential amalgamation of Building Information Modeling (BIM) and machine learning methodologies for automated property valuation presents encouraging outcomes. However, it is important to acknowledge that the study’s scope may be constrained by the accessibility and reliability of BIM data and the intricate nature of real-world situations that our models may not comprehensively encompass. Furthermore, the precision and dependability of the valuation models may be impacted by the particular datasets employed for training and validation.

2.2 Decision-making models for realty management

Hu and Lin [19] designed a multi-criteria group decision-making methodology (MCGDM) for property management systems under a z-number environment. The suggested technique is a novel aggregation method that decreases the complexity of management systems. MCGDM identifies the problems during decision-making that reduce the latency in the computation process. The suggested MCGDM increases the decision-making process’s accuracy, enhancing the systems’ performance range. The practical implementation of the proposed integrated multi-criteria group decision-making methodology for property concealment risk assessment may face challenges due to the intricate nature of collecting dependable and precise Z-number data, encompassing both fuzzy and probabilistic uncertainties. Furthermore, the efficacy of the methodology may be impacted by the proficiency and subjective assessments of the decision-makers engaged in the collective decision-making process.

Syed and Lawryshyn [20] introduced a multi-criteria decision-making technique for the risk analysis process. Important key values and Risk Matrix (RM) are identified from the database, producing appropriate data for further decision-making. A cost-benefit analysis is utilized here to analyze the appropriate data required for the systems. RM produces accurate datasets, which reduce the complexity and latency in the risk assessment process. The introduced method reduces the risk issues, increasing the network efficiency and effectiveness. Due to its complexity and processing overhead, the suggested multi-criteria decision-making approach may hinder its practical applicability in real-time or time-sensitive scenarios.

Fu and Hu [21] proposed a binary tree pricing method using a machine learning (ML) technique for farmland management systems. A strategy gradient reinforcement learning algorithm is implemented in the pricing technique to analyze the important features of the database. The proposed method uses basic principles which produce feasible conditions for the analysis process. Compared with other approaches, the proposed technique increases the financial criteria and performance of the networks. However the method consumes high computation complexity.

Lee et al. [22] introduced an adaptive decision-making method for civil infrastructure systems. The primary objective of the introduced technique is to evaluate the risks that occur during infrastructure systems. The introduced method identifies the dynamic changes in the communication and interaction process. Community risks and issues are identified, providing feasible data for decision-making. The introduced method achieves high accuracy in decision-making but consumes high computation time.

Anejionu et al. [23] developed a cloud-enabled big data infrastructure for spatial urban data systems (SUDS). The developed method uses an analytic technique that analyzes SUDS’ important characteristics and features. Variations of urban keys and values are identified from the database, decreasing the computation process’s latency. The real-time data source provides relevant data for the analytics process. Simulation outcomes illustrate that the suggested technique increases prediction accuracy, which improves SUDS’s performance range. However, the method requires optimization approach while handling large volume of data.

Zhang and Carter [24] proposed a grazing system model for industries. The proposed model is an online information system that provides various services and functions to the users. Remote sensing techniques are used here to capture sensible data using wireless sensors. Wireless sensors reduce the difficulty of the computation process. The objective of the recommended model is to identify climate change in grazing lands. The recommended model enhances the overall effectiveness ratio of grazing land systems. Although, the training system requires additional effort to manage the scalability.

Zhang et al. [25] developed a data-driven decision strategy for offshore assets. A multicore classification model is utilized here that recognizes the exact information types based on the conditions. The data mining method is also used here to detect important datasets for decision-making. The multicore classification model improves offshore assets’ feasibility and computation efficiency. The developed model maximizes the accuracy in decision-making, enhancing the systems’ efficiency and reliability range.

März [26] introduced a new decision-making method for small private landlords (SPL). Both theoretical and practical aspects of the data, which produce relevant data for various processes, are identified. SPL gathers information from the low-demand housing market, providing feasible data for decision-making. Certain privacy policies and schemes are used in SPL, enhancing the systems’ performance range. The presented method attains high accuracy in decision-making, high the computation process’s complexity range.

Teresa [27] developed a racialized class-monopoly rent for land contracts. The primary objective of the proposed technique is to offer proper security services to the owners. The suggested technique identifies the vulnerabilities and issues that occur during rent. The identified effects provide relevant data for security policies. The proposed method creates various services via the class-monopoly process. The suggested approach maximizes the security and privacy range in the renting process. Although requires the standard security techniques to ensure the privacy on large volume of data.

Peng [28] proposed a machine-learning assessment method for hazard shock. The key objective of the suggested technique is to determine property and land prices for the decision-making process. Disaster notification and alert messages that provide important characteristics and features for various processes are analyzed. Key features of hazard effects and shock are also identified, reducing the computation process’s latency. The suggested technique increases the performance and efficiency level of the systems.

A study by Zhou et al. [29] proposed using AI to aid decision-making in the great data age. The study begins by surveying the relevant articles from the International Journal of Information Management (IJIM) to provide a historical perspective of AI. Following this, the article delves into AI’s decision-making capabilities in general and the challenges of integrating and interacting with AI to supplement or replace human decision-makers. This article offers twelve research proposals to IS researchers concerning conceptual and theoretical development, AI technology-human interaction, and AI implementation. These proposals aim to promote the study of AI decision-making in the Big Data age.

Foster Provost and Tom Fawcett presented data science, big data, and data-driven decision-making [30]. Conceptual frameworks, which are an integral aspect of data science, facilitate this. Automated pattern extraction from data, for instance, follows a standard procedure. By breaking down problem-solving into its component parts, one may make it more methodical, less error-prone, and easier to understand. Big data technology, data science methods grounded on big data, and data-driven decision-making have a strong track record of significantly boosting company performance.

Kaile Zhou et al. [29] suggested using big data for smart energy management. A four-pronged approach is covered: demand-side management (DSM), asset management and collaborative operation, power generation-side management, and microgrid and renewable energy management. After that, we’ll look at the evolution of smart energy management inside industries powered by big data and talk about it. In conclusion, the author stresses the difficulties of smart energy management powered by big data in information technology infrastructure, data gathering and governance, data integration and sharing, processing and analysis, privacy and security, and qualified individuals.

The data analytics-based technique for strategic decision-making was introduced by Murat Ozemre and Ozgur Kabadurmus [31]. This research uses a large quantity of publicly available trade data to predict future export quantities using two distinct machine learning algorithms: Random Forest (RF) and Artificial Neural Networks (ANN). The BCG Matrix, used for strategic market analysis, incorporates the predicted values. A made-up case study of a Chinese firm that exports freezers and refrigerators is used to verify the suggested technique. Based on the findings, it is clear that the suggested approach facilitates efficient strategic market research and produces reliable trade predictions. Regarding the accuracy of forecasts, the RF outperforms the ANN as well.

Business strategy alignment and big data analytics capability (BDAC) were both addressed by Akter [32] as means to enhance company performance. The BDAC model is suggested in this work by combining the entanglement perspective of sociomaterialism with the resource-based theory (RBT). Using an entanglement conceptualization of the higher-order BDAC model and its effect on FPER, the results validate its usefulness. The findings also show how analytics capability-business strategy alignment significantly moderates the BDAC—FPER connection.

The relationship between big data analytics and the performance of firms was discussed by Samuel Fosso Wamba et al. [33]. This study adds to the existing literature by investigating the impact of business process architecture on firm performance (FPER) and how process-oriented dynamic capabilities (PODC) mediate this connection. The results confirm the importance of the hierarchical BDAC model’s entanglement conceptualization, which directly and indirectly affects FPER. The findings back up PODC’s substantial mediation function in boosting FPER and insights. The author discusses what this means for future studies and clinical work.

Big data analytics’ strategic business value was proposed by VARUN GROVER [34]. By building on previous work on the value of IT, this research provides a framework for understanding BDA, and it then uses real-world examples of BDA to show how this framework works. Then, we review the framework’s capabilities in studying BDA value generation and realization-focused components and linkages. The author takes a problem-oriented approach to the framework, arguing that issues with BDA components could spark specific concerns for further investigation. Research and practice in BDA may be better targeted via the strategic use of data resources if this study is structured in a way that allows for the development of a substantial research agenda.

Organizational decision-making might benefit from the Human-AI Symbiosis, as proposed by Mohammad Hossein Jarrahi [35]. Humans can still provide a more comprehensive, intuitive approach to handling ambiguity and uncertainty in organizational decision-making, but AI may augment human cognition when it comes to complicated problems via increased computational information processing capability and an analytical approach. The concept of "intelligence augmentation" is similar to this premise: artificial intelligence technologies should complement human work, not replace it.

Healthcare Big Data and the Medical Internet of Things were proposed by Dimiter V. Dimitrov [36]. The MIoT is essential to healthcare’s digital transformation because it paves the way for new business models, facilitating process modifications, productivity gains, cost reduction, and improved customer experiences. These days, exercise, health education, symptom monitoring, care coordination and collaborative illness management are all made possible with smartphone applications and wearable technology. Users will spend less time making sense of data outputs thanks to platform analytics that make data interpretations more relevant. Discoveries made possible by analyzing massive amounts of data will propel the digital revolution in healthcare, corporate operations, and real-time decision-making.

Implementing big data analytics and business intelligence was pioneered by Hsinchun Chen et al. [37]. As a result of the severity and breadth of data-related issues facing modern businesses, business intelligence and analytics (BI&A) has grown in prominence among academics and industry professionals alike. Using almost a decade of relevant academic and business publications, the author presents the results of a bibliometric analysis of influential BI&A articles, scholars, and study subjects. At last, the six papers that make up this special issue are presented and described according to the BI&A research concept.

Based on the survey, there are several problems with existing methods in attaining a high processing rate, including stagnancy detection, success rate, training ratio, and reduced processing time. Hence, this study introduces a knowledge-based processing scheme (KPDS) for validating real estate property data based on market value, maintenance, and sales criteria.

The dataset from [38] validates the proposed scheme’s performance. This data is acquired from various real estate systems and their logs. The rental, sales, maintenance, and mortgage data are validated through 18 classes, 4931 entries, and 323894 events. Therefore, the first analyses are presented to justify the need for KDPS. Fig 1 presents the average maintenance expenses between 2014 and 2022 for rental and sales.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Property analysis (Source: Self).

https://doi.org/10.1371/journal.pone.0307043.g001

Fig 1 representations reflects the property components of reality management item description, monthly rental, amenities, unit area with building floor count, size and location. Indicators of the financial condition of these properties include the amount of the deposit and the monthly rent, while specifics such as the number of rooms, the unit space, and the number of floors in the building provide information on the buildings’ physical characteristics. It is also possible for the existence of amenities such as elevators and storage facilities to have an impact on the popularity and value of the aforementioned properties.

Using context-aware information, users may better comprehend the recommendations and their relevance to the larger realty management objectives. Developing user-friendly interfaces that let people investigate and query the model’s decision-making process. Users may better understand the underlying variables and examine particular recommendations with the help of interactive tools. Incorporating domain-specific information and experience into the data analysis process is a limitation addressed by the knowledge-dependent data processing scheme (KDPS). The utilization of human expertise and decision-making rules allows the system to manage limitations in data and surmount computational intricacies effectively. The versatility of KDPS enables its application in several industries, effectively overcoming the challenge of limited generalizability through integrating industry-specific knowledge and data sources.

It is recommended that the KDPS be improved by integrating real-time data streams from diverse sources, including property listings, market trends, and economic indicators. The constant integration of real-time data into the system makes the analytical and decision-making processes grounded in the most current information. Furthermore, it is recommended that the KDPS incorporate methods to facilitate prompt updates and model retraining, enabling the system to adjust swiftly to evolving market conditions. One potential approach is to implement periodic or event-driven updates, wherein the knowledge base and decision rules are revised in accordance with the most recent data and market conditions. By overcoming this constraint, the KDPS may offer more precise and pertinent observations, empowering real estate experts to make well-informed choices that accurately represent the present condition of the market.

4.1 Data processing rate

The data processing rate is high in this method, which uses the knowledge-learning technique to improve the values. The data processing is done based on the knowledge determined from the amenities, sales, and data reports. This process is done on two levels; at first, the data will be selected for analysis, and the second level will be used to improve the values of the process. The enhancement and the stagnancy will be checked during this data processing, and further steps will be taken to reduce the flaws. The output of the cumulative gross data also helps in the processing levels of the data and eliminates the flaws during stagnancy. The output of the process will be verified with the market value to determine its state and whether it is improved or not. If the values are increased, then they are enhanced with good maintenance. If the values are reduced, it has flaws due to the poor amenities, and some steps should be taken to reduce the flaws. Recommendations are given to increase the upcoming data processing values rather than the market values without stagnancy (Fig 9). Feedback should be promptly obtained following interactions to get insights while the user speculates about the experience. The insights derived by real-time or rapid feedback systems are more likely to be correct and relevant.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 9. Data processing rate.

https://doi.org/10.1371/journal.pone.0307043.g009

4.2 Stagnancy detection

The stagnancy detection is great in this technique by the learning technique in the processing levels. Based on the outcome of the process, the stagnancy can be determined, and then suggestions can be made to enhance the upcoming realty process. Property stagnation may be influenced by external variables that might change over time, such as neighbourhood development, municipal legislation, and economic situations. The challenge with relying on historical data alone to predict stagnation is that it may not fully capture or forecast these changes. If there is low maintenance amenity and poor sale information, stagnancy occurs in the realty procedure. Stagnancy occurs when the realty process is not maintained better regarding its amenities. The knowledge of data reports helps determine the stagnancy in the procedure, and during this stagnancy period, the necessary training is provided to enhance the process. Based on the stagnancy outcome, the upcoming new reality process will be improved with more efficacious training given to the data. In the stagnancy period, the data are verified for the processing levels, and if there is any lag, then the needed training will be provided to the process to eliminate flaws. Based on the sales and amenity knowledge, the flaws can be detected, and further steps are taken to eliminate the flaws in the upcoming realty management processes (Fig 10).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 10. Stagnancy detection.

https://doi.org/10.1371/journal.pone.0307043.g010

4.3 Processing time

The time taken for processing is less in this method, which utilizes the data processing scheme to enhance the values. The outcome of the cumulative big data helps in the processing levels of the data and checks the stagnancy in the process. From the realty process, the information on the sales, amenities, and data reports are extracted to form the gross information for the successful process. Data processing uses knowledge from production, sales, and previous process outcomes. Further new processing for improvements and flaw mitigation is recommended based on the knowledge determined from successful production and gross data outcome. There should be frequent maintenance of the real process and the required facilities for a better sale. This ensures definite operation of the manufacturing process, meets the real-time requirements and eliminates stagnancies. Effectively processing large-scale data over a cluster of devices is made possible by adopting distributed computing frameworks. As data volumes increase, this KPDS method permits horizontal scalability by introducing more processing nodes to the environment. The sales can also be increased by improving the maintenance of the industrial process and the facilities of the realty process (Fig 11).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 11. Processing time.

https://doi.org/10.1371/journal.pone.0307043.g011

4.4 Success rate

This method’s success rate is better using the processing technique to improve the realty process values. The cumulative data at different stages are used to improve production and delivery efficiency. Big data has large amounts of information originating from high amounts of information gained from data reports, amenities, and sales, which needs new processing techniques to enable better process expansion. Gross cumulative data is used for processing levels to reduce flaws and stagnancy and enhance present values. The market value will be referred to for the verification process. The data selection will be done using knowledge-learning techniques based on the previous stagnant outcomes of this process. This input may also help enhance the upcoming realty management process. This output enables the upcoming process results without the shorter processing time flaws. In this processing procedure, all the cumbersome and inefficiency can be detected and eliminated. Performing bias audits to examine past biases’ effect on model predictions systematically. This study analyzes the model’s performance across different demographic groupings and pinpoint potential sources of bias for the decision-making process. In this way, the success rate is high in this method (Fig 12).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 12. Success rate.

https://doi.org/10.1371/journal.pone.0307043.g012

4.5 Training ratio

This method’s training ratio is high, and it uses knowledge learning to enhance real-time data processing. After detecting the stagnancy in the process, the needed training is provided to the upcoming process to result in a success rate when compared with the market value. If the output of the verification process is high, then the market value will be such that there will be no need for further training. If the outcome is less than the market value, then training will be given to the upcoming realty processes. A recommendation is given after the market value has been verified. A recommendation is provided to overcome the stagnancy and to improve the realty process value. This is used to enhance the data processing levels to achieve market value and eliminate the flaws in the procedure. Manifest data operation is done to eliminate the flaws of the first realty procedure. This helps to increase the processing ratio, success ratio, and stagnancy reduction. The real estate market is unpredictable and dynamic, with underlying trends that may shift over time. The model could have trouble modifying if the environment changes substantially while the KDPS still assumes stagnant circumstances. The training ratio is high in these techniques with a knowledge learning scheme (Fig 13). Tables 1 and 2 highlight the proposed schemes’ performance from the above comparison. Strong security and privacy protocols must be implemented to deal with the delicate nature of real estate data. Building trust with consumers and complying with privacy requirements requires protecting data’s confidentiality, integrity, and availability.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 13. Training ratio.

https://doi.org/10.1371/journal.pone.0307043.g013

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Highlights of performance assessment (data inputs).

https://doi.org/10.1371/journal.pone.0307043.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Highlights of performance assessment (amenities).

https://doi.org/10.1371/journal.pone.0307043.t002

4.6 Implication of this study

According to the results, the knowledge-dependent data processing scheme (KDPS) demonstrates superior performance compared to the other techniques, specifically SMMP, BIM+AVM, and BTPM, in many real estate management measures. KDPS has the maximum processing rate of 253, which is much greater than the other approaches. On the other hand, SMMP has the lowest processing rate of 59. The KDPS strategy has the largest percentage of cases where the process becomes stagnant or stuck, measuring 42.312%. This indicates that KDPS is more effective than the other techniques in identifying and handling sluggish situations. KDPS exhibits the shortest processing time, measuring 2.153 seconds, indicating its exceptional efficiency and quickness in managing realty activities. KDPS approach exhibits the highest success rate, as evidenced by its proportion of effective outcomes at 0.844 or 84.4%. This indicates that KDPS outperforms the other methods in generating successful results. The training ratio, which quantifies the number of repetitions or iterations needed per input, is observed to be the highest for KDPS at 111. This suggests that achieving ideal performance may necessitate further training or exposure to data. The findings indicate that the knowledge-dependent data processing scheme (KDPS) demonstrates efficacy and efficiency in the context of realty management tasks. It surpasses alternative methods in multiple metrics, such as processing rate, stagnancy detection, processing time, success rate, and training ratio.