Research on Optimization Design of Talent Cultivation Path in Tourism Industry Based on Data Mining under Vocational Education System

We are living in the information age, and the explosion of data has created an urgent need for powerful techniques to help us find valuable information from massive amounts of data. This need has led to the birth of data mining, which is the process of extracting interesting patterns and knowledge from large amounts of data. Data mining is an emerging technology, unlike traditional statistical analysis, data mining has a huge amount of workload and a great deal of unknowns in the mining results, so it is often used to reveal important information that is undiscovered and valuable in a large amount of data [21].

Generally speaking, the process of data mining includes three stages: data preprocessing, data mining, and result evaluation. The data mining process is shown in Figure 1. 1)

Data pre-processing stage

As the data stored in the database of the computer system has problems such as noise, missing data, and redundancies, it is not suitable for direct data analysis and mining. Therefore, the need for further preprocessing of raw data, and the quality of preprocessing directly affects the quality of data mining results. The data preprocessing stage requires data cleaning, integration, generalization, and transformation, and the final preprocessed data should have the characteristics of accuracy, completeness, and consistency.

Figure 1.

Data mining process

In order to accurately describe the association rule mining problem and facilitate the discussion of the problem, a formal definition of the association rule mining problem needs to be given. The following is a definition of the association rule mining

Definition 1 The dataset mined by an association rule is denoted as D (D is the transaction database), D={t₁,t₂,…,t_k,…,t_n}, t_k = {i₁,i₂,…,i_j,…,i_p}, (k = 1,2,…,n), is a transaction. The element i_j(j = 1,2,…,p) in t_k is called Item.

Definition 2 Let I={i₁,i₂,…i_m} be the set of all items in D, any subset X of I be called the itemset in D, and |X|=k be called the set X the itemset in k. Let t_k and X be the transaction and itemset in D, respectively, and if X⊆t_k, say transaction t_k contains itemset X.

Transactions and itemsets, although both are collections of items, have different meanings. A transaction is a constituent element of database D (similar to a record or tuple in a relational database), while an item is simply a combination of items (similar to a field in a relational database) specified for the purpose of extracting association rules [23]. The containment relationship between a transaction and a set of items indicates that the items in the set are related to each other for that transaction.

Definition 3 The number of transactions in dataset D that contain itemset X is called the support number of itemset X, denoted as σ_x. The support of itemset X, denoted as sup port(X), is the probability of P(X). 1support(X)=σx| D |×100%

Where: |D| is the number of transactions in dataset D. If sup port(X) is not less than the user specified minimum support (notation: minsupport), then X is called a frequent itemset (or large itemset), otherwise X is called an infrequent itemset (small itemset).

Definition 4 Let X and Y be the itemsets in data set D. 1)

If X ⊆ Y, then 2support(X)≥support(Y)

Definition 5 If X, Y are itemsets, and X ∩ Y = Φ, implication X ⇒ Y is called the association rule, X, Y are called the premise and conclusion of association rule X ⇒ Y, respectively. The support of item set X ∪ Y is called the support of association rule X ⇒ Y, and is the percentage of affairs in D that contain X ∪ Y, i.e., the probability P(X ∪Y), denoted as: sup port (X ⇒Y ). 3support(X⇒Y)=support(X∪Y)=P(X∪Y)

The confidence level of association rule X ⇒ Y is the percentage of things in D that contain X that also contain Y, i.e., the conditional probability P(Y|X), denoted confidence (X ⇒ Y). 4confidence(X⇒Y)=support(X∪Y)support(X)×100%=P(Y|X)

Usually, the minimum confidence level specified by the user according to the extraction needs is denoted as minconfidence.

Support and confidence are two important concepts for describing association rules, the former is used to measure the statistical importance of an association rule in the whole dataset, and the latter is used to measure the degree of confidence of an association rule. Generally speaking, only association rules with high support and confidence are likely to be interesting and useful for users.

Usually, users specify the minimum support (denoted as minsupport) and the minimum confidence (denoted as minconfidence) according to the extraction needs. The former describes the minimum importance of an association rule, and the latter specifies the minimum reliability that an association rule must satisfy.

Definition 6 e.g. sup port(X ⇒ Y)≥ min sup port e.g. confidence (X ⇒ Y) ≥ m min confidence calls association rule X ⇒ Y a strong rule, otherwise it calls association rule X ⇒ Y a weak rule [24].

The problem of association rule mining is to solve all association rules in D whose support and confidence exceed minsupport and minconfidence respectively, i.e., to solve rule X ⇒ Y which satisfies sup port(X ⇒ Y)≥min sup port and confidence (X ⇒ Y)≥min confidence.

An example is shown in Fig. 2, where the classification of I = {Footwear, Shoes, HikingBoots, Clothes, Outerwear, Jackets, SkiPants, Shirts} and I is shown.

Figure 2.

The classification of 1

The logarithmic average salary is shown in Figure 4. At present, the salary of professionals in the tourism industry is still relatively high in general, mostly concentrated in the range of 8K-20K. However, since there are many human-caused outliers in the raw salary, all salary analysis will be based on the average salary after the commonly used logarithmic processing. The average salary is calculated by dividing the highest and lowest salaries offered for the job. The graph shows that the logarithmic mean salary is roughly evenly distributed, indicating that the job outlook for travel industry professionals is still stable and impressive.

Figure 4.

Logarithmic mean salary

The size of the company is closely related to the economic strength of the company, so the analysis of the size of the company will also help the data tourism industry talents to find jobs with better employment prospects. The box plot of salary versus company size is shown in Figure 5. The larger the company size, the higher the corresponding logarithmic average salary, especially for some foreign-funded and joint ventures. This is also releasing a signal to the tourism industry talents, in the process of choosing a career to learn more about the size and culture of the company, as a prerequisite to screen out the employment opportunities offered by quality companies. Small and medium-sized companies data more demand for tourism industry talents, while the demand of large companies with more than 5,000 employees is less. This practical conclusion may not be in line with the theoretical assumptions, but it does provoke in-depth thinking. Although on the surface of the small and mediumsized enterprises for the tourism industry talent to provide more employment opportunities, but whether it reflects the current tourism industry-related companies are mostly small and medium-sized enterprises in the initial development stage, large enterprises due to their own structural characteristics, all kinds of talent reserves are more complete, but on the contrary, the demand for tourism industry talent is less than small and medium-sized enterprises.

Figure 5.

Salary and company scale plan

The accumulation of work experience is still particularly important for those who work in the field of tourism industry, and work experience should be accumulated at least to a certain stage in order to have a more complete and independent working mind. The box plot of average salary versus work experience is shown in Figure 6. As a rule, the salary promised by employers increases with the number of years of work experience they require. However, since the travel industry does not require too much work experience, the main focus is on 3-4 years or less. There is not much difference in salary levels for those with 2 years of work and below, but there is a significant increase in salary for those with more than 5 years of work experience, and there is no significant increase in salary after the required number of years of work experience reaches 7 years or above.

Figure 6.

Average salary and work experience box diagram

The association relationship between courses and courses is obtained through the association rule experiment, and the association analysis between courses is shown in Table 3. In the table, it can be found that the rule Introduction to Tourism, Hospitality and MICE A1 (1-1) = C20 => Introduction to Tourism, Hospitality and MICE A2 (1-2) B = 19, and failure to study Introduction to Tourism, Hospitality and MICE A1 will affect the grade of Introduction to Tourism, Hospitality and MICE A2. Based on the data in the table, it can be inferred that if the rules Introduction to Tourism, Hospitality and MICE and Destination Management are not studied seriously, it will affect the graduation design grade, and the related correlation results later in this paper show that the graduation design will have a direct impact on the individual’s work. The seventh data shows that if there are problems in the economic analysis of tourism industry in the second semester of freshman year and food culture and food service management in the second semester of sophomore year, it will lead to unsatisfactory grades in the first semester of junior year on tourism consumer behavior. The results of the association rules obtained above indicate that some courses are related to each other in some way, they affect each other, and the good or bad grades of some courses will directly affect the grades of the following courses. Under the credit system, schools can use these potential rules to guide students’ learning and teaching.

The correlation between courses and internships is shown in Table 4. Based on the data in the table, it can be inferred that unsatisfactory grades in Club Management and Destination Management will affect the internship in the fourth year. Whereas, courses such as Tourism Consumer Behavio Introduction to Landscape Architecture, Tourism Project Management, Hospitality Managemen Information Systems, Tourism Economics, Intermediate Financial Accounting, and Sociology o Tourism are helpful for the senior year internship.

The correlation analysis between courses and Offer is shown in Table 5. From the table, it can be concluded that Tourism Consumer Behavior (1-1) = Tourism Economics (2-2) = A12 => Offer = Y12, which can be inferred that if Tourism Consumer Behavior and Tourism Economics are not studied seriously, it will affect the eligibility to get an offer.

The analysis of the association between courses and high salary is shown in Table 6. Tourism Consumer Behavior, Tourism Economics, Tourism Geography, Tourism Psychology, Rural Tourism Planning and Design, Community Tourism Planning and Management, Revenue Management, Destination Management, Introduction to Landscape Architecture, Introduction to Tourism, Hospitality, and MICE, Big Data and Smart Tourism, Intermediate Financial Accounting, Food and Catering Culture and Restaurant Management, and Tourism Destination Management have an impact on the ability to obtain Offer. Intermediate Financial Accounting, Frontier Issues in Tourism, Introduction to Software Engineering, Economic Analysis of the Tourism Industry, Tourism Planning and Development, Tourism Geography, Tourism Consumer Behavior, Rural Tourism Planning and Design, Architectural Photography, and Revenue Management have a critical impact on the ability to achieve a high salary.

In summary, Tourism Geography, Food Culture and Catering Management, Intermediate Financial Accounting, Big Data and Smart Tourism, Introduction to Tourism, Hospitality and MICE, Destination Management, Tourism Consumer Behavior, Revenue Management, Introduction to Landscape Architecture, Rural Tourism Planning and Design, Tourism Economics, Community Tourism Planning and Management, Tourism Psychology, Tourism Industry Economic Analysis, Architectural Photography, Tourism Economics are the most important courses that have a vital impact on the courses that have an impact on the jobs and salaries of graduates.

Breaking the inherent curriculum structure, reconstructing the tourism curriculum system with professional group thinking, realizing professional integration, culture and tourism fusion and skill interaction, and upgrading the knowledge structure and professional skills simultaneously. Restructure the knowledge structure with cross-border integration thinking, highlighting generalized teaching, cross-border integration, and the transformation of traditional knowledge into intelligent ability. Design the teaching content with the thinking of integration of books and certificates, and introduce the work standards of “National Tourist Guide Qualification Examination”, “Design and Implementation of Study Tour Curriculum” and “Tourism Customizer” to provide effective choices for students to master a variety of skills. Provide effective options for students to master various skills. We design the practical links with the mindset of integration of production and education, and provide progressive training for the skills of tour guide, travel customization, and study guide, so as to realize the complementary enhancement of the digital skills of “Internet+Tourism”.

Docking the industry development and the new needs of tourism consumption, re-establish the knowledge, ability composite cross-border intelligent tour guide personnel training specifications. In the process of cultivation, it is necessary to penetrate the Internet, intelligent and new media knowledge, construct a composite knowledge system, realize the transformation of basic service skills to composite skills such as the application of digital technology, new media marketing, and intelligent service, and realize the cultivation goal of iterative upgrading of knowledge structure and ability structure. At the same time, it is necessary to strengthen students’ cross-border thinking, intelligent thinking, and innovative thinking, and enhance their scientific and technological innovation literacy.

Cultivate cross-boundary and composite tour guide talents through professional integration and industry-teaching fusion, carry out inter-industry and inter-professional cultivation of tour guide talents from a borderless perspective, and form a professional group by integrating related professions, so as to make the knowledge system, the curriculum system, and the practical teaching system integrated and coherent. At the same time, deepen the school-enterprise cooperation, alternating work and study cultivation mechanism, build a new tourism talent incubation base, through the learning cognitive stage, learning and training apprenticeship stage, skills enhancement stage, project combat stage, internship stage, comprehensive enhancement stage, forming a complete alternating work and study talent cultivation chain, prompting the transformation of static teaching into dynamic teaching, the transformation of single-skill learning into composite cross-border skill learning, and the innovation of the teaching process to realize both the quality of talent cultivation and the quality of employment, and to realize the quality of employment. Realize that there is a double improvement in the quality of talent training and employment.

Establishing new standards for employment positions, coordinating and standardizing training materials and teachers, etc., and launching a selection process for social tourism training institutions, focusing on supporting and building training bases adapted to new forms of tourism and new occupations. At the same time, relying on virtual reality, human-computer interaction, big data and other advanced information technology, breaking the demand for training venues to create immersive training base to meet the diverse learning needs of different students, to promote the “Internet +” “smart tourism” education new form.