The reuse of industrial and mining wasteland is the concrete embodiment of improving the carrying capacity of resources and environment. It is also an urgent requirement for the smooth transformation and high-quality development of resource-based cities. Tourism reuse is one of the main uses of industrial and mining wasteland, and the related research mainly embodies in ecological restoration policy [1, 2], ecological restoration [3, 4], ecological landscape reconstruction [5,6,7,8], industrial genetic protection and cultural tourism resource development. Among them, the protection of industrial and mining cultural heritage and the development of cultural tourism resources are called ‘post-industrial landscape design’ [9, 10]. The studies are summarized as follows.
The first study is the research of ecological restoration policy. The US government promulgated the
The second study is the research of ecological restoration engineering technology. Application of ecological restoration engineering technology spans topics such as ecological reconstruction species selection and configuration [3], ecological restoration models and evolution patterns [4, 13], evaluation of land suitability for reclamation [14] and land reclamation and ecological reconstruction research methodologies [15,16,17,18,19]. Focussing on restoration of a diamond mine landscape in Namaqualand, South Africa, Carrick and Krüger [20] suggested that the greatest challenges to restoration stem from the unsuitability of much of the mined overburden soils for plant growth and recommended use of a combination of various ecological techniques for soil improvement. Similarly, Boruvka et al [7] compared the effects of different soil restoration techniques on open-cast mine reclamation. The influence of natural conditions and visual and landscape effects on the reclamation of derelict land has also been researched.
The third study is research of post-industrial landscape design [9, 10, 21,22,23]. In 1970, Richard Haag, a famous American landscape architect, transformed a heavily polluted 8-hectare wasteland near the Port of Seattle into a popular park through pollution control. Similar projects include Citroen Park (1970) in Paris, France, and Gas Works Park (1972–1975) in Seattle, the US, both of which have become classic examples of post-industrial landscape design. In the 1980s, industrial heritage tourism began to become popular in Europe. A great number of industrial heritage sites in Britain, Norway, Germany, Sweden, and other countries have since been included in the World Heritage List, further motivating research and development of derelict land tourism. Post-industrial landscape design theory in Europe developed gradually in the 1990s, when industrial heritage protection was promoted. As a result, a large number of post-industrial landscape parks, including the Golpa-Nord Opencast Mine Park in Germany, the Byxbee Park in the US and the Biville Quarry Park in France, were constructed.
The industrial tourism reuse mode of industrial and mining wasteland is still in the exploratory practice stage in China. In terms of system, fund, supervision and operation, it has not yet formed a more effective and sustainable safeguard mechanism. The problems are mainly embodied in two aspects: (1) non-systematic in pre-planning and (2) insufficient means of recommendation.
To solve the above problems and increase public awareness of tourist attractions, some researchers have already applied recommendation systems to tourism. Currently, recommendation systems are generally classified into graph- [24,25,26,27], trust-relationship- [28, 29] and matrix factorization-based [30, 31] systems, although other alternatives also exist. Collaborative filtering methods are applied, which are based on user similarity measured by cosine similarity or Person correlation. The major problems encountered in collaborative filtering are data sparsity and cold-start issues [32], which can significantly lower the recommendation accuracy. Graph-based recommendation is a relatively flexible collaborative filtering method, which can solve the data sparsity problem. Previously, Shen et al. [33] proposed a tourist attraction recommendation system based on the situational context and trust relationship. This algorithm, which is based on collaborative filtering, solves the data sparsity problem by replacing the user similarity with the user trust relationship. In addition, introduction of a situational context to recommendation systems can reflect the personalized demands of users in a more comprehensive manner. Borras et al. [34] have considered a smart e-tourism recommendation system, which can detect features related to tourist attractions through user mobile devices and then provide users with useful information such as weather forecasts, touring routes and site business hours. Recommendation systems applied to tourism suffer from various problems, especially cold start and data sparsity. This is because users do not typically visit tourism websites as frequently as online shopping or other websites; thus, it is more difficult for a recommendation system to collect the historical user data and the preferences of the majority of users.
In this paper, a method that combines the experience economy theory with a recommendation algorithm based on user profile is proposed. The main contributions of this paper are the following statements:
We use experience economy tools and propose 20 indicators of five implementation strategies so as to integrate the surrounding tourism resources and target the comprehensive benefits of regional tourism resources. The overall goal is to guide the planning of industrial tourism to achieve a suitable marketing strategy. User profile technology is used to establish a model of different users. Based on this model, tourists’ fuzzy and potential demands are identified to address the problem of cold-start items in the recommendation process. Accurate recommendation is achieved in the operation stage. A list of recommended tourist attractions is generated using the user-profile-based recommendation system. This is accomplished by obtaining labelled data on the users. The recommendation systems we proposed run precisely on a sparse dataset. The rationality of applying the proposed method in the field of industrial tourism and the recommendation system accuracy are verified through experiments performed on real datasets.
The industrial wasteland reuse model proposed in this paper is established according to the following steps: (1) Tourist data are collected to model user behaviour; then, through procedures such as label grading, user modelling, clustering and label abstraction, user profiles are constructed; (2) industrial tourism planning is performed based on user profile tags and user characteristics extracted from questionnaires based on experience economy; (3) a tourist attraction recommendation system based on user profiles and random walks is constructed considering the tourist user profiles and behaviour characteristics.
User profiles are descriptions of certain groups of people created by abstracting tags based on analysis of demographic information, travel preferences, social relationships and consumption behaviour. User profile construction requires establishment of a label grading system based on the business conditions. In accordance with the demographic principles and business conditions considered in this study, the labelling system is divided into the following three levels: level 1, which includes demographics, content, behaviour and business attributes; level 2, which includes location, economic strength, travel preference and brand label information, and level 3, which includes age, gender, home address, purchasing power, travel equipment, eating habits, consumption habits and activeness information.
First, user interest models (UIMs) are constructed. Unique identifiers
Users are then clustered through a clustering algorithm. Users with similar UIMs are utilized to form a user profile, while users with significant variations in the UIMs are assigned to different user profiles. The clustering algorithm proposed in this paper is similar to the
First, a cluster centre
Experience economy is based on life and situation, shaping sensory experience and thinking identity, influencing and understanding users’ consumption behaviour. As a major feature of experience economy, ‘relationship’ emphasizes the long-term relationship with customers. Enterprises should consolidate and develop the relationship through repeated transactions so as to grasp the characteristics of users. Based on this, we combine the experience economy implementation tools proposed by Schmitt and user profiles, aimed to offer useful advice for development strategies for industrial wasteland tourism development.
It was found that the experience economy implementation tools applied to industrial tourism mainly encompass visual signs, media, service process, personnel characteristics, spatial surroundings and external associations. The population characteristics related to industrial tourism are as follows: there are more men than women among tourists aged 15–25 and 50–60 years; these individuals are middle- and high-income earners; and they have college degrees or higher.
In this study, a questionnaire survey was designed to investigate industrial tourism experience economy in Handan. The survey was administered at five typical industrial tourism sites in Handan. From 500 respondents, 469 (93.8%) completed and returned the questionnaire appropriately. This close-ended questionnaire contained two sections, experience economy combination and industrial tourism awareness, and had 20 questions in total. The 5-point interval scale was employed, with 5 points indicating ‘strongly agree’ and 1 point indicating ‘strongly disagree’.
The retrieved questionnaire data were analysed using IBM SPSS Statistics V22.0. Principal component analysis was employed, and the eigenvalues of four factors were extracted, explaining 60.17% of the population variance. The results are listed in Table 1.
Experience economy factor analysis.
Dimensions | Variables | Factor loading | |||
---|---|---|---|---|---|
1 | 2 | 3 | 4 | ||
Visual signs | Tourism signs | 0.839 | |||
Theme posters | 0.716 | ||||
Media | Advertising | 0.661 | |||
News coverage | 0.643 | ||||
Hotspot-based marketing | 0.502 | ||||
Website promotion | 0.543 | ||||
Service process | Sightseeing | 0.809 | |||
Object display | 0.419 | ||||
Knowledge distribution | 0.511 | ||||
Interactive experience | 0.798 | ||||
Product purchases | 0.344 | ||||
Personnel characteristics | Tour guides | 0.598 | |||
Service staff | 0.661 | ||||
Visitors | 0.398 | ||||
Spatial surroundings | Landscape features | 0.801 | |||
Infrastructure | 0.702 | ||||
Supporting facilities | 0.698 | ||||
Neighbouring attractions | 0.809 | ||||
External associations | Government support | 0.344 | |||
Industry alliance | 0.399 | ||||
Tourism cooperation | 0.609 |
Table 1 indicates that the 20 variables can be attributed to 4 factors. Apart from object display, visitors, product purchases, government support and industry alliance, the loading values of all other factors were >0.5, indicating that the questionnaire had good convergence.
The data were then subjected to multiple linear regression analysis; the results are presented in Table 2.
Multiple linear regression.
Variable | Unstandardised coefficient | Standardised coefficient |
Sig | ||
---|---|---|---|---|---|
Standard error | |||||
Tourism signs | 0.180 | 0.045 | 0.213 | 3.163 | 0.000 |
Theme posters | 0.054 | 0.056 | 0.060 | 0.861 | 0.197 |
Advertising | 0.183 | 0.048 | 0.254 | 3.697 | 0.000 |
News coverage | 0.051 | 0.060 | 0.062 | 0.852 | 0.397 |
Hotspot-based marketing | 0.173 | 0.045 | 0.213 | 3.612 | 0.000 |
Website promotion | –0.026 | 0.040 | –0.049 | –0.838 | 0.401 |
Sightseeing | 0.003 | 0.049 | 0.005 | 0.003 | 0.997 |
Object display | 0.015 | 0.051 | 0.022 | 0.312 | 0.744 |
Knowledge distribution | 0.071 | 0.045 | 0.091 | 1.191 | 0.197 |
Interactive experience | 0.189 | 0.049 | 0.201 | 3.510 | 0.001 |
Product purchases | –0.026 | 0.047 | –0.063 | –0.547 | 0.689 |
Tour guides | 0.016 | 0.058 | 0.019 | 0.276 | 0.897 |
Service staff numbers | 0.033 | 0.061 | 0.046 | 0.568 | 0.546 |
Visitors | –0.020 | 0.062 | –0.023 | –0.317 | 0.752 |
Landscape features | 0.081 | 0.061 | 0.097 | 1.340 | 0.175 |
Infrastructure | 0.027 | 0.058 | 0.035 | 0.452 | 0.675 |
Supporting facilities | –0.034 | 0.048 | –0.052 | –0.703 | 0.483 |
Neighbouring attractions | 0.181 | 0.046 | 0.223 | 3.299 | 0.000 |
Government support | 0.084 | 0.047 | 0.125 | 0.918 | 0.071 |
Industry alliance | 0.035 | 0.057 | 0.046 | 0.658 | 0.510 |
Tourism cooperation | 0.008 | 0.060 | 0.011 | 0.133 | 0.895 |
From Table 2, the
As a result of the poor recognition of industrial tourism and the small number of tourists, the major problems facing development of a recommendation system for industrial tourism products are data sparsity and cold starts. In this study, a recommendation system called recommendation-user-tag-project (R-UTP) is proposed, which is based on a tripartite graph and a random walk algorithm. The definition of the user-tag-project (UTP) relationship tripartite graph is given below.
A UTP graph is a directed tripartite graph (
The UTP graph can be represented by two modified bipartite graph adjacency matrixes:
After the UTP graph is constructed, the characteristics of the nodes in the UTP graph are combined and the new R-UTP recommendation system is designed on the basis of random walks. Here, we take the
Here,
Next, we extend the above resource diffusion method to the UTP tripartite graph. For the
Finally, the resource diffusion method is applied to the UTP tripartite graph and linear blending is performed between
If the target user
To verify the feasibility of the plan and the accuracy of the recommendation system, a validation experiment was performed. This experiment was divided into two parts: (1) experiments on hit ratio and accuracy on real-world data and (2) verification of the development strategy based on user characteristics obtained from tourism statistics.
Experiments were performed on the standard dataset of the world's largest travel website, TripAdvisor, and the CiteULike dataset. Handan is a province-controlled city in the south of Hebei Province, China. It lies between the Taihang Mountains to the west and the North China Plain to the east, bordering the Shanxi, Shandong and Henan provinces (Figure 2). Known as the city of iron and coal, Handan is estimated to have 480 million and 4 billion tons of these two resources, respectively.
Experiments were performed to evaluate the algorithm performance and compare it with similar algorithms. Each dataset could be divided into test and training sets, that is, 80% of the data were used as the training set and the remaining 20% were used as the test set. The experimental environment was the Windows 7 operating system with an Intel i5 3.3G CPU and 8GB memory.
In this study, the hit ratio and recall ratio were used to measure the effectiveness of the proposed algorithm. First, a list of the top-
Correspondingly, the recall ratio refers to the ratio of projects of interest to the user against the projects of interest to all users in the system, as measured by Eq. (6), where |
The algorithm proposed in this paper was compared with two classical algorithms: the user-based system filtering algorithm UserKNN and the object-based collaborative filtering algorithm ItemKNN. First, the hit ratio variation in accordance with different values of
The R-UTP algorithm and the other two algorithms were executed on the two datasets individually. The experimental results are shown in Figure 2. With the change of
The recall ratio experiment was also performed on the TripAdvisor and CiteULike datasets, with the recall ratios obtained for different |
Figure 3 shows the change in recall ratio in accordance with the number of recommended projects |
In addition, experiments on the novelty of recommendation were also performed. The novelty of recommendation refers to the maximum possible recommendation of products that are not well known to users, in addition to recommendations based on tourist interests. As industrial tourism is poorly recognized by the public at present, novelty should be an important parameter in the proposed recommendation system. The recommendation novelty was calculated as follows:
The experimental results regarding the novelty of recommendation are shown in Figure 4. It is apparent that increasing
Industrial tourism has been proven to play a significant role in ecological environment improvement and landscape restoration in areas surrounding industrial and mining wastelands and can also promote local economic growth. Industrial tourism, combining the industry and tourism, is a new type of tourism integrating sightseeing, leisure, nostalgia, experience, investigation, learning and shopping. It is the outcome of increasingly diversified tourism demands against the backdrop of ongoing urban and economic transformation. The experience economy theory derived from the experiential economy provides a feasible solution to the problems faced by industrial tourism. Thus, industrial tourism can become a new driver of economic growth by meeting the new demands of tourists, with tourism resource integration, development of in-depth tourism products and use of innovative marketing methods.
In future, tourism product development should be further supported and diversified, and the tourism resources around abandoned industrial and mining areas in resource-based cities should be integrated. This will help realize sustainable development of industrial tourism and provide a reference for protection of industrial cultural heritage and development of a cultural tourism industry in mining areas. Hence, construction of a green and safe transformation model for resource-based cities will be promoted.