Utilizing multi-source data in popularity prediction for shop-type recommendation
Introduction
It is an important but challenging task for an investor to determine the most appropriate shop type for a given location [1]. Shop-type recommendation aims to suggest an optimal shop type (such as restaurant, cafe, and bar) in a given commercial location to attract more consumers. Selecting a suitable shop type is not only beneficial to reducing investment risks, but also crucial for the development of community economy and future business [1].
One of the key problems for shop-type recommendation is predicting shop popularity which can be represented by the number of attracted consumers [1], [2]. Features, which are closely related to shop popularity, can be categorized into two classes: the external characteristics (e.g., geographical environment [3], people flow [4] and commercial structure [5]) and the internal characteristics (e.g., environment [6], service [7] and product quality [8]). The impacts of these features vary on different shop types. For example, accessibility is more important for a convenience store while good service is more important for a cafe. In the real-world, investors usually make decisions based on their subjective judgments and perceptions on these features. Due to insufficient information and the limitation of cognition, it is difficult for investors to quantify and distinguish the impacts of multiple features. This may lead to flawed decisions and thus increase the subsequent investment risks. Therefore, it is important to collect sufficient information concerning these features and propose an effective approach to determine the suitable shop type for investors.
During the last decade, the rapid development of location-based services and big data mining technologies have made it possible to collect and analyze multi-source data from various platforms for decision support. Business review sites, such as Dianping.com,1 contain shop-related information and customer ratings. Location-based navigation systems, such as Ditu.amap.com,2 provide information on numerous points of interest (POI) and their relevant information. Moreover, mobile carriers collect pervasive spatio-temporal trajectories of people’s daily movement. With these data sources, the aforementioned shops’ external and internal characteristics can be evaluated. In this paper, we utilize these data to support the recommendation. One example of the data is depicted in Fig. 1. However, after obtaining such data, there is still a challenge during the recommendation: how to provide decision support of shop-type recommendation by integrating multi-source heterogeneous data.
In order to address this challenge, we propose a shop-type recommendation method integrating multi-source information to predict shop popularity for shop-type recommendation. First, we collect multi-source data from business review sites, location-based navigation systems and mobile carriers and preprocess the data using the MapReduce framework implemented with the Hadoop platform. Subsequently, we construct a location-type matrix by aggregating shops’ popularity to alleviate the problem of incomplete data. Then, the location profile and the commercial structure are constructed from multi-source data to describe the location’s features at micro and macro levels. In addition, a hybrid similarity measurement is proposed to measure the closeness between locations with the consideration of the location profile and the commercial structure. Finally, on the basis of the location-type matrix and the hybrid similarity, a location-based collaborative filtering is adopted to predict shop popularities for all shop types. For a given shop location, the top- shop types with the highest predicted popularities are recommended to investors.
The main contributions of our work can be summarized as follows.
First, multi-source information and big data analytics are utilized to study the shop-type recommendation problem. In comparison with traditional ways of decision making based on investors’ subjective judgments and perceptions, the proposed method considers more relevant factors and provides reliable decision support for investors to overcome cognitive limitation and reduce investment risks.
Second, location profiles are constructed from multiple aspects by considering internal and external characteristics simultaneously, which provides a comprehensive description of locations. Particularly, internal characteristics are analyzed in this problem for the first time and help to characterize location profiles more accurately.
Third, a hybrid similarity model integrates the location profile and the commercial structure into a unified framework. The location profile and the commercial structure are used to measure the similarity between locations at micro and macro levels. The combination of the two perspectives contributes to improving recommendation performance.
Fourth, the principle of collaborative filtering is introduced to deal with the shop-type recommendation problem for the first time, where a location-type matrix is proposed to alleviate the problem of data incompleteness. The proposed method takes full advantage of commercial information of other locations to provide valuable references for recommending a suitable shop type to the target location.
Finally, the performances of the proposed method are validated through comparison with several benchmark methods on a real-world dataset. This dataset is collected from a business review site, a location-based navigation system and a mobile carrier, related to 29673 shops, 10965 locations and 17 million individuals in Beijing, China.
The remainder of this paper is organized as follows. Section 2 reviews the background and related works. Section 3 describes the problem and introduces the research framework. Section 4 describes the process of recommendation generation in detail. The experiments and results analysis are discussed in Section 5. Finally, Section 6 gives a conclusion including our contribution and the direction of future work.
Section snippets
Location-based recommendation systems
Due to their powerful capability in solving information overload, recommender systems have attracted a lot of attention in the past decades. Many recommendation problems have been widely studied [9], [10], [11], [12], [13]. With the rapid development of intelligent mobile devices and information technologies, many recent studies have explored data-driven recommendation problems. Location-related data is the main data source of recommendation, which has been explored by many scholars. Quercia
Problem statement
This paper proposes a novel shop-type recommendation method that suggests suitable shop types in a given location for a shop investor. Let be a set of locations, and be a set of candidate shop types, such as a beverage shop, restaurant or dessert shop. We use to represent the popularity of shop type for location , which can be represented as the number of attracted customers [1]. We aim to generate a ranking of predicted popularities for all shop types
Recommendation generation
To generate recommendations, a location-based collaborative filtering approach is adopted to suggest suitable shop types. According to the principle of collaborative filtering, a location-type matrix and the similarity between locations are necessary [29]. First, we construct the location-type matrix using data from the business review site and the location-based navigation system. Then we propose a hybrid similarity measurement, combining the location-profile-based similarity and the
Data collection and preprocessing
We evaluate the proposed method based on a real-world dataset, which is collected from three platforms, including Dianping.com, Ditu.amap.com and a mobile carrier. Dianping.com is one of the largest websites containing consumer ratings on shops in China, which can provide basic shop information. Ditu.amap.com is one of the largest navigation systems in China, which can provide basic location information. The mobile carrier can provide anonymous cell-id trajectory data, which is used to analyze
Discussion and conclusion
In this study, we propose a shop-type recommendation method integrating multi-source information to suggest suitable shop types in a given location for investors. Specifically, the location-type matrix is constructed to alleviate the incomplete data problem. Then, the location profile and the commercial structure are constructed from multi-source data to describe the features of the location at micro and macro levels. A hybrid similarity measurement is proposed to measure the similarity between
Acknowledgments
This work is supported by the National Natural Science Foundation of China (Grant No. 91746111, Grant No. 71672140), Ministry of Education & China Mobile Joint Research Fund Program (No. MCM20160302), supported also by the Spanish Ministry of Economy and Competitiveness with FEDER funds in the grant the grant TIN2016-75850-R.
References (59)
- et al.
Shop-type recommendation leveraging the data from social media and location-based services
ACM Trans. Knowl. Discov. Data.
(2016) - et al.
Retail store location selection problem with multiple analytical hierarchy process of decision making an application in Turkey
Procedia-Social Behav. Sci.
(2012) - et al.
The impact of an exciting store environment on consumer pleasure and shopping intentions
Int. J. Res. Mark.
(2015) - et al.
Retail service quality and perceived value: A comparison of two models
J. Retail. Consumer Serv.
(1997) - et al.
A google wave-based fuzzy recommender system to disseminate information in university digital libraries 2.0
Inform. Sci.
(2011) - et al.
Hiperion: A fuzzy approach for recommending educational activities based on the acquisition of competences
Inform. Sci.
(2013) - et al.
A semantic enhanced hybrid recommendation approach: A case study of e-Government tourism service recommendation system
Decis. Support Syst.
(2015) - et al.
A cross-domain recommender system with consistent information transfer
Decis. Support Syst.
(2017) - et al.
A trust-semantic fusion-based recommendation approach for e-business applications
Decis. Support Syst.
(2012) - et al.
A social route recommender mechanism for store shopping support
Decis. Support Syst.
(2017)
User identification for enhancing IP-TV recommendation
Knowl.-Based Syst.
A multi-latent transition model for evolving preferences in recommender systems
Expert Syst. Appl.
Collaborative filtering with ordinal scale-based implicit ratings for mobile music recommendations
Inform. Sci.
The effect of retail store environment on retailer performance
J. Bus. Res.
Pedestrian flow characteristics studies: A review
Transp. Rev.
Socotraveler: Travel-package recommendations leveraging social influence of different relationship types
Inf. Manage.
Enhancing collaborative recommendation performance by combining user preference and trust-distrust propagation in social networks
Knowl.-Based Syst.
Utilizing customer satisfaction in ranking prediction for personalized cloud service selection
Decis. Support Syst.
Member contribution-based group recommender system
Decis. Support Syst.
A web recommendation system considering sequential information
Decis. Support Syst.
One-class collaborative filtering based on rating prediction and ranking prediction
Knowl.-Based Syst.
Recommendation systems: A probabilistic analysis
J. Comput. System Sci.
Factored similarity models with social trust for top-n item recommendation
Knowl.-Based Syst.
Personalized recommendations based on time-weighted overlapping community detection
Inf. Manage.
Integrating AHP and data mining for product recommendation based on customer lifetime value
Inf. Manage.
SCoR: A synthetic coordinate based recommender system
Expert Syst. Appl.
Modelling socially-influenced conditional preferences over feature values in recommender systems based on factorised collaborative filtering
Expert Syst. Appl.
Geo-spotting: Mining online location-based services for optimal retail store placement
The role of store location in influencing customers’ store choice
J. Emerg. Trends Econ. Manag. Sci.
Cited by (10)
mHealth App recommendation based on the prediction of suitable behavior change techniques
2020, Decision Support SystemsCitation Excerpt :With the development of information technology, an increasing volume of individuals' data can be collected and extracted [26, 27]. These data serve as an important source of information for logging the daily behavior of individuals [28], such as telephonic communication behavior, cell-phone usage behavior, and activity-travel behavior [29-32]. In this study, we describe the BCT-related characteristics of users and integrate them with BCT to provide personalized recommendations by using data collected from one of China's largest mobile operators.
A complementing preference based method for location recommendation with cellular data
2019, Knowledge-Based SystemsCitation Excerpt :We discuss the experiments in Section 5 and offer concluding remarks in Section 6. Recommender systems have been widely studied for their ability to help users cope with the problem of information overload [19–23]. Among them, location recommendation has drawn a great deal of attention for its practical application, such as improving users’ experience in a city.
Community-based Dynamic Graph Learning for Popularity Prediction
2023, Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data MiningRecommendation Model Combining Implicit Influence of Trust with Trust Degree
2022, Moshi Shibie yu Rengong Zhineng/Pattern Recognition and Artificial IntelligenceAn ensemble of deep learning algorithms for popularity prediction of flickr images
2022, Multimedia Tools and ApplicationsApplication Research of Personalized Recommendation Technology in College English Teaching Reform under the Background of Big Data
2021, Proceedings - 2021 6th International Conference on Smart Grid and Electrical Automation, ICSGEA 2021