A study of the role of Internet technology in promoting innovative and entrepreneurial activities in school education

With the continuous development and application of science and technology, a large number of new technologies and new business models have emerged, which require talents to have more comprehensive and advanced skills and literacy in order to adapt to the development and changes in society. Innovation and entrepreneurship education is one of the effective ways to cultivate such high-quality talents. Innovation and entrepreneurship education in colleges and universities under the perspective of “Internet+” can not only provide a good entrepreneurial environment and support, but also guide students to be based on the actual needs, and create more innovation and entrepreneurship opportunities by providing a full range of innovation and entrepreneurship services and resources [1-3]. This helps to attract more entrepreneurial talents and promote innovation and entrepreneurship activities within colleges and universities, as well as to promote more communication and cooperation between colleges and universities and society [4-5]. Innovation and entrepreneurship education in colleges and universities based on Internet technology is of great significance for building an innovative country and promoting the development of globalized economy [6]. Entrepreneurship and innovation is becoming the theme of economic development on a global scale, and the innovation and entrepreneurship education in colleges and universities, as an important position of innovation and entrepreneurship, will play a more important role [7-8].

The era of “Internet+” is an era full of changes and opportunities, and the innovation and entrepreneurship education model of college students should keep pace with the times, update the education content and education methods, pay attention to the development trend of the times and future market demand, and help students grasp the market opportunities and future development direction [9-12]. Internet technology breaks the limitations of the traditional education model and provides students with open and shared innovative and entrepreneurial education resources without geographical restrictions, so that students can easily access resources and experiences in the global context, promote international cooperation and innovation, and broaden their horizons and ideas [13-16]. Internet technology also provides students with a wealth of learning tools to achieve personalized learning paths and open resource sharing, which helps students break the boundaries of disciplines and fields, and promotes interdisciplinary and cross-field innovation [17-19]. In addition, the innovation and entrepreneurship education model in the era of “Internet+” also encourages students to actively use digital tools and technologies to carry out practical activities, emphasizes the combination of theoretical knowledge and practical application, and develops students’ skills in the field of innovation and entrepreneurship [20-22].

Under the Internet technology, this paper constructs an innovation and entrepreneurship resource platform based on personalized recommendation from the demand and function design of innovation and entrepreneurship resource platform. After that, the data in the innovation and entrepreneurship resource platform is calculated by similarity, and then the similarity between students’ ability and utility is evaluated from the six labels of students’ “theoretical ability, application ability, competition experience, leadership ability, collaboration ability and energy value”. Based on the evaluation results, we recommend appropriate innovation and entrepreneurship resources for students. Finally, we evaluate the effectiveness of the model and the satisfaction after application.

2

Construction of an innovation and entrepreneurship resource platform based on personalized recommendation

2.1

Construction of Innovation and Entrepreneurship Resource Platform

2.1.1

Objectives of the platform

The construction goal of the personalized recommendation-based university innovation and entrepreneurship resource base system is to provide a resource sharing and rapid independent learning platform for university students participating in innovation and entrepreneurship projects. The users of the system are university students participating in innovation projects, teachers guiding innovation projects, enterprise personnel interested in university students’ innovation projects, and managers of the resource library. In order to meet the above objectives, the system will perform the following functions: 1)

The resources managed by the resource library system include documents, pictures, codes, videos, texts and other project resources, and at the same time, it is able to provide the function of video preview, so that the learners can preview the relevant project resources online without having to download them to the local area, which improves the user learning experience.

2)

The resource library system provides different access rights for different users, and the rights of different types of users are independent and do not interfere with each other, so the system must have a clear rights management system.

3)

The resources contained in the resource library are massive and increasing, in order to enable learners to quickly find the learning resources they need, the system needs to provide a variety of query methods, this paper, according to the characteristics of the innovative project resources, provides a full-text search, keyword retrieval, project name search, project category search four kinds of retrieval, the retrieval of resources to provide comments, scoring and download functions.

4)

The resource library system is based on the architecture, everything is displayed in the form of page and the interface design should strive to be beautiful, easy to use, simple, and can have clear operating tips to enhance the learners’ experience.

5)

The system can provide rich learning resources, the quantity and quality of resources is an important index to measure the good or bad of the resource library system, a good system can realize the benign interaction with the learners, the learners can upload good resources to the resource library, the resource library can provide learners with high-quality learning resources, so as to maintain the real-time nature of the library’s resources and increase the learners’ interest in using it.

6)

The resource base can provide learners with personalized learning resources, change the information pulling service of the traditional resource base into an active push service, and actively display the resources that users are interested in in front of the users, so as to improve the timeliness, effectiveness and relevance of the system in disseminating learning resources.

2.1.2

Analysis of the platform’s user requirements

Based on personalized recommendation algorithm [23-24], the innovation and entrepreneurship resource base system for college students is divided according to the roles as students participating in the innovation project, teachers guiding the innovation project, enterprise personnel interested in the innovation project, and system administrators. Different roles have different business requirements, and different recommendation methods are used for different households. The following business use cases for different users are presented.

The repository administrator business use case is shown in Figure 1:

Examples of instructor use are shown in Figure 2:

The student example is shown in Figure 3:

An example of an operational staff use case is shown in Figure 4:

2.1.3

Analysis of platform resource requirements

The construction of the resource system of the innovation and entrepreneurship resource base system for college students based on personalized recommendation is to organize the resources of the resource base with certain clues, which is convenient for learners to query and learn. The resource library is based on the item as the basic organizing unit, and the smallest retrieval unit is the code, document, picture, video and other materials, the item has multiple attributes, and the materials are also divided into multiple types. First of all, there are multiple attributes such as item level, discipline classification, college classification, instructor, project incubability, and project application area in college students’ innovation and entrepreneurship items. There are many types of resources in the personalized recommendation-based college student innovation and entrepreneurship [25] repository, but the resources can be divided into two major categories, text and video, by extracting the commonalities of the resources. In this paper, openo ffice+swftools+flexpaper technology is introduced to realize the unified preview of text resources.

2.1.4

Platform interface requirements

The system interface is the medium of interaction between the user and the system, and all the operations of the user on the system are carried out on the interface. A friendly and easy-to-use interface can make the user work more accurately, efficiently and pleasantly, and can give full play to the performance and characteristics of the system, and the needs of the interface of the university student innovation and entrepreneurship resource base system based on personalized recommendation are as follows: 1)

Page content: the content needs to highlight the theme, the terminology and font format is standardized and unified, and the overall layout is beautiful and reasonable.

2)

Navigation structure: the page should have eye-catching navigation instructions, which is easy for users to understand and convenient for operation.

3)

Technical environment: the page can adapt to the commonly used various browsers, and can meet the different resolution browsing.

4)

Page style: the page is full of beauty, content and style, terminology and page font size and color to be coordinated.

5)

Page operation: page operation is simple, according to the user needs to be able to respond at any time to the problem of user help.

2.1.5

Analysis of the functional structure of the platform

Through the analysis of user business use cases in the section, the functional diagram of the university student innovation and entrepreneurship resource base system based on personalized recommendations can be derived. The functional structure diagram of the system is shown in Figure 5. The six functional modules included in the system are shown below:

2.2

Personalized recommendation technology based on student innovation and entrepreneurship competition

2.2.1

Resource similarity algorithms

For each resource G_j,1 ≤ j ≤ N_G,N_G is the number of innovation and entrepreneurship competitions. The competition model is represented by a hexadecimal group 〈Gpf,Gcl,Gta,Gpa,Gce,Gre〉, which represents the field of specialization of the competition, the rating of the competition, the theoretical ability and application ability examined by the competition, the competition experience gained from participating in the competition, and the fatigue value consumed. By modeling the competition G and creating a vector of competition attributes by extracting the competition attributes, the similarity between the competition G_j1 and the competition G_j2, simi(G_j1,G_j2) is shown in the following equation [26]: (1) $\begin{matrix} s i m i (G_{j 1}, G_{j 2}) = 1 - \sqrt{+ {(G t a (G_{j 1}) - G t a (G_{j 2}))}^{2} + {(G t a (G_{j 1}) - G t a (G_{j 2}))}^{2}} \\ + {(G c e (G_{j 1}) - G c e (G_{j 1}))}^{2} + {(G r e (G_{j 1}) - G r e (G_{j 1}))}^{2} \end{matrix}$

2.2.2

Student model construction based on personalized recommendation

For each student S_i,1 ≤ i ≤ N_s, N_s is the number of students. The student model uses a hexadecimal group 〈Sta,Spa,Sce,Sla,Sca,Sae〉 to represent the labels of their abilities and utilities, which represent the student’s theoretical ability, application ability, competition experience, leadership ability, collaboration ability, and energy value.

1)

Student’s ability model

The theoretical competence Sta(S_i) of student S_i is shown by the following equation: (2) $S t a (S_{i}) = \frac{\sum_{h = 1}^{N_{T C} (S_{i})} (ω_{t g} \times t g (S_{i}, C_{h}) + ω_{t e} \times t e (S_{i}, C_{h})) \times t c (C_{h})}{\sum_{h = 1}^{N_{T C} (S_{i})} t c (C_{h})}$

Where N_π(S_i) denotes the number of theory courses attended by student S_i; tg(S_i,C_k) denotes student S_i grade in theory course C_k:

te(S_i,C_k) denotes the evaluation of student S_i classroom performance in the theory course C_k, which is determined by the student’s attendance, completion of class assignments, etc.; and tc(C_k) is the credit corresponding to the theory course C_k: ω_zz and ω_kε weighting parameters, ω_zz + ω_tε = 1.

Student s_i ability to use Spa(S_i) is shown in the following equation: (3) $S p a (S_{i}) = \frac{\sum_{k = 1}^{N_{p c} (S_{i})} p g (S_{i}, C_{k}) \times p c (C_{k})}{\sum_{k = 1}^{N_{p c} (S_{i})} p c (C_{k})}$

where N_PC(S_i) denotes the number of lab courses in which student S_i participated; and pg(S_i,C_h) denotes student S_i grade in lab course C_k; pc(C_k) is the credit equivalent of the lab course C_k.

The competition experience Sce(S_i,G_ji) of student S_i in competition G_j1 is shown in the following equation: (4) $S c e (S_{i}, G_{j 1}) = \frac{\sum_{j = 1}^{N_{g (i)}} (G c e (G_{j 2}) \times s i m i (G_{j 1}, G_{j 2}))}{N_{G (S_{i})}}$

where N_G(5) denotes the number of competitions in which the student has participated. Gce(G_j2) indicates the race experience of the race G_j2. simi(G_j1,G_j2) indicates the similarity between race G_j1 and race G_j2.

Student S_i Leadership Competency Sla(S_i) and Collaboration Competency Sca(S_i) are shown in the following equation: (5) $S l a (S_{i}) = \frac{\sum_{k = 1}^{N_{(s, t)}} a v g (E l a (S_{i}, T_{k}))}{N_{(S, T)}}$ (6) $S c a (S_{i}) = \frac{\sum_{k = 1}^{N_{(x, x)}} a v g (E c a (S_{i}, T_{k}))}{N_{(S, T)}}$

Where N_(S,T) represents the number of teams in which student S_i participated; avg(Ela(S_i,T_k)) represents the average leadership evaluation of student S_i in team T_k; and avg(Eca(S_i,T_k)) represents the average collaboration evaluation of student S_i in team T_k.

The energy value Sre(S_i) for student s_i is shown in the following equation: (7) $S r e (S_{i}) = 1 - \sum_{j = 1}^{N_{o (s_{j})}} G r e (G_{j})$

where N_{G(S_i)} denotes the number of students S_i participating in Competition G; Gre(G_j) denotes the fatigue value of Competition G_j.

2)

Student utility model

When a student participates in a competition, he/she will gain a corresponding benefit, which is defined here as “utility”. Corresponding to the six attributes of students, the student utility model is also represented by a hexadecimal group 〈Uta,Upa,Uce,Ula,Uca,Uae〉, which represents the theoretical ability utility, application ability utility, competition experience utility, leadership ability utility, collaboration ability utility, and energy value utility of students.

Student S_i Participation in Competitions G_j theoretical competence utility represents the gains in theoretical competence that a student can make when participating in competitions, as shown in the following equation: (8) $U t a (S_{i}, G_{j}) = S t a (S_{i}) \times G t a (G_{j})$

Where Sta(S_i) represents the theoretical competence of student S_i and Gta(G_j) is the theoretical competence attribute of competition G_i Student S_i Utilized Ability Utility to Participate in Competitions G_j represents the gains in utilized ability that a student can achieve when participating in competitions, as shown in the following equation: (9) $U p a (S_{i}, G_{j}) = S p a (S_{i}) \times G p a (G_{j})$

Where Spa(S_i) represents the theoretical competence of the student S_i and Gpa(G_j) is the utilized competence attribute of the competition G_i. Student S_i Participation in Competitions G_i Competition Experience Utility represents the benefit that a student can gain in terms of competition experience when participating in competitions, as shown in the following equation: (10) $U c e (S_{i}, G_{j}) = S c e (S_{i}, G_{j}) \times G c e (G_{j})$

Where Sce(S_i,G_j) represents the experience value that student S_i can earn on Competition G_j and Gce(G_j) is the experience value attribute for Competition G_j.

Student S_i Participation in Competitions The leadership utility of G_j represents the gains in leadership that a student can make while participating in competitions, as shown in the following equation: (11) $U l a (S_{i}, T_{k}) = S l a (S_{i}) \times E l a (T_{k})$

Where Sla(S_i) is the leadership competence of student S_i and Ela(T_k) denotes the leadership competence rating value of all student members in team T_k as shown in the following equation: (12) $E l a (T_{k}) = \frac{a v g (S l a (T_{k}))}{1 + \frac{1}{var (S l a (T_{k}))}}$

Among them, avg(Sla(T_k)) indicates the mean value of the leadership ability of all students in team T_k, the higher the value means the better the leadership ability of the team; var(Sla(T_k)) indicates the variance of the leadership ability of all students in team T_k, the higher the value means the higher the difference of the leadership ability of the team members, and it is easy to form a more harmonious team.

The Collaborative Ability Utility for Student S_i to participate in the competition G_j represents the gains in collaborative ability that a student can achieve while participating in the competition, as shown in the following equation: (13) $U c a (S_{i}, T_{k}) = S c a (S_{i}) \times E c a (T_{k})$

Where Sca(S_i) denotes the collaborative ability of student S_i; Eca(T_k) denotes the value of collaborative ability rating of all students of team T_k as shown in the following equation: (14) $E c a (T_{k}) = \frac{a v g (S c a (T_{k}))}{1 + var (S c a (T_{k}))}$

Among them, avg(Sca(T_k)) indicates the mean value of the collaboration ability of all students in team T_k, and the higher the value means the better the collaboration ability of the team; var(Sca(T_k)) indicates the variance of the collaboration ability of all students in team T_k, and the smaller the value means the smaller the difference of the collaboration ability among students, which is easier to work together.

The energy value utility for student S_i to participate in competition G_j represents the benefit that the student can gain in terms of energy value while participating in the competition as shown in the following equation: (15) $U a e (S_{i}, G_{j}) = \sin (S a e (S_{i}) \times \frac{π}{2})$

In this equation, Sae(S_i) is the energy value of the student S_i situation sine function limits the utility of the value range of [-1, 1], when the energy value is too low or negative, it means that the student participated in too many competitions resulting in a lack of energy, which will cause a negative impact affecting the competitions that have been participated in.

Based on the above utility model, the utility of student S_i to participate in competition G_j is shown in the following equation: (16) $U (S_{i}, G_{j}) = {(\begin{matrix} ω_{i a} \times U t a (S_{i}, G_{j}) + ω_{ρ a} \times U p a (S_{i}, G_{j}) + \\ ω_{o c} \times U c e (S_{i}, G_{j}) + ω_{o c} \times U a e (S_{i}, G_{j}) \end{matrix})}^{\begin{matrix} α \times U l a (S_{i}, T_{k}) + \\ (1 - α) \times U c a (S_{i}, T_{k}) \end{matrix}}$

Where ω_w,ω_pw,ω_ce,ω_ae is a weighting parameter, 0 ≤ ω_ta,ω_pw,ω_ce,ω_ae ≤ 1 , and ω_iα+ω_pα+ω_cò+ω_aò = 1, for balancing the weights of theoretical ability utility, utilization ability utility, competition experience utility, and energy value utility; α is a weighting parameter, 0 ≤ α ≤ 1, for balancing the weights of leadership ability utility and collaboration ability utility.

The total utility of Team T_k is shown in the following equation: (17) $U (T_{k}) = \sum_{S_{i} \in T_{i}} U (S_{i}, G_{j})$

2.2.3

Optimization objectives and constraints

Based on the above analysis, the optimization objective and constraints are shown in the following equation to maximize the utility of the team and its members by selecting the right students to form a team to participate in the right competition: (18) $\begin{matrix} \max U (T_{k}) \\ \max U (S_{i}, G_{j}) \\ s . t . S_{i} \in T_{k} \\ 0 \leq i \leq N_{S} \\ 0 \leq j \leq N_{G} \\ 0 \leq k \leq N_{T} \end{matrix}$

3

Analysis of the application effect of the personalized resource recommendation platform

3.1

Results and Analysis of Personalized Recommendation Algorithm

In this experiment, the dataset is divided into two parts: a test set and a training set to ensure experiment accuracy. The experiment is carried out five times, one data set is selected as the experimental test set in turn, and the remaining four data sets are used as the experimental training set, using the 5-layer cross-validation method, and the average of the results of the five experiments is taken as the final evaluation value.

3.1.1

Simulation results and analysis

When performing simulation analysis to study the impact of each variable on the overall system, it is often taken to adjust the value of a variable to observe its specific impact on the system. The design of personalized recommendation systems for innovation and entrepreneurship platforms often requires feature labels for student ability and utility, where the number of label groups represents the number of defined student labels. In this paper, when a student is tagged with 3, 5, and 6 tags respectively, we explore the relationship between the resources recommended by the personalized recommendation-based innovation and entrepreneurship repository system and the student’s matching degree after 100 days with different number of tags.

The results of the effect of time on the matching of personalized recommended resources with different number of labels are shown in Figure 6. It can be seen that when the number of labels to students is 3, the resources recommended by the personalized recommendation algorithm peaks only at the 50th day, and from the 50th to the 100th day, the match between the resources recommended by the personalized recommendation-based innovation and entrepreneurship repository system and the students maintains around 59%, with an overall low degree of match. When the number of students is 5, the matching degree reaches the peak at about 23 days, and when the experiment is carried out to the 100th day, the matching degree basically stays at about 84%, and the overall matching degree is relatively high. When the number of labels given to the student is 6, the initial match between the recommended resources and the student is about 68%, the match grows rapidly to about 90% from day 1 to about day 7, and the match between the personalized recommended resources and the student stabilizes at 95% when on day 100. Comprehensive analysis reveals that as the number of labels on students increases, the recommended resources have a higher match with students and it takes less time to reach a high match. This shows that within a reasonable range, the resources recommended under the number of explicit feature labels can have a more positive impact on students.

3.1.2

Accuracy of Personalized Recommendation Algorithms

Three personalized recommendation methods are used to import 3,568 innovation and entrepreneurship resource information, and five experiments are conducted. Set the recommendation method proposed in this paper as PRBSCE, and in the control method, the content-based recommendation algorithm is CBRA, and the content-based collaborative filtering algorithm is CBCFA. The comparison results of the accuracy rate of the test under different algorithms are shown in Fig. 7. The results show that the accuracy rate of each algorithm is relatively high when the number of recommended resources is small, and decreases as the number of recommended resources increases. From the figure, it can be seen that the accuracy rate of this paper’s recommendation algorithm is significantly higher than that of collaborative filtering algorithm and content-based recommendation algorithm, and the accuracy coefficients of the three algorithms are 0.9839, 0.8264, and 0.6555, respectively, when the recommended resource is 1. With the increase in the number of resources, the decreasing trend of the recommendation algorithm proposed in this paper is relatively moderate compared to other algorithms. And the experiment found that when the recommendation of PRBSCE algorithm is in the range of 2-6, its accuracy coefficient is in the range of 0.8942-0.9183, which is a very small difference. And the accuracy coefficients of the other two algorithms are between 0.4475-0.6841 and 0.4126-0.5724 respectively, with a large difference. This is because the model of innovation and entrepreneurship repository system based on personalized recommendation constructed in this paper is able to provide timely feedback and update according to the change of user’s interest, whereas the CBRA and CBCFA algorithms lead to a larger decrease due to the sparse matrix. This indicates that the recommendation algorithms proposed in this paper perform better than other algorithms when the data is sparse.

3.2

Application and Evaluation of Personalized Recommendation Platforms

In order to further verify the effectiveness of the application of the personalized recommendation platform in this paper in promoting innovation and entrepreneurship activities in schools, this paper guided students to practice. The target of the practice is the students of two classes studying the course “Innovation and Entrepreneurship Education and Practice” in the first semester of the 2023-2024 academic year in School A. There are 300 students in the class. After the system was applied, this paper conducted a questionnaire survey on the junior students who had used the system. A total of 300 questionnaires were distributed and 300 valid questionnaires were recovered, with a valid recovery rate of 100%. The mean value of Cronbach’s Alpha coefficient of the questionnaire is 0.8853, which indicates that the consistency of the questionnaire is good, so the results of this survey have excellent reliability. Next, this paper analyzes the results based on the students’ satisfaction with the system and their satisfaction with the learning outcomes.

3.2.1

Interface Design Satisfaction Results and Analysis

This paper investigates the satisfaction with the interface design of resource recommendation platforms. Investigations were carried out in four aspects: ①“The interface design is unique, the structure is clear, and the main board interfaces are arranged in an orderly manner”. ②“The interface style is unique, in line with aesthetics, habits, experience is very good”. ③“The navigation menu is designed to be reasonable, convenient, familiar, and can be quickly adapted to”. ④“The interface design closely meets the clear needs of learning, entrepreneurship and other Demand, targeted, can quickly find the target resources”. Satisfaction was categorized into five dimensions, namely, “A Very Compliant, B Compliant, C Mostly Compliant, D Not Compliant, and E Very Not Compliant”. The results of the interface design satisfaction survey are shown in Table 1. The results show that, ① The distribution of the aspect samples is dominated by “very compliant” and “compliant”, with frequencies of 117 and 68, accounting for 39% and 22.67% respectively. ② The frequency is 132.48, accounting for 44%. ③ The distribution of the aspect samples is dominated by “very compliant”, “compliant” and “mostly compliant”, accounting for 42.67%, 22.33% and 24.67% respectively, with corresponding frequencies of 128, 67 and 74. The corresponding frequencies are 128, 67 and 74, respectively. ④ The distribution of the aspect samples is dominated by “general conformity”, with a frequency and percentage of 137 and 45.67%, respectively. In summary, the respondents gave high ratings in terms of satisfaction with the interface design, which is well-designed.

Table 1.

Interface design satisfaction survey results

Evaluation content	Level of satisfaction	A	B	C	D	E
①	Frequency	117	68	36	52	27
①	Percentage(%)	39	22.67	12	17.33	9
②	Frequency	132	76	61	29	2
②	Percentage(%)	44	25.33	20.33	9.67	0.67
③	Frequency	128	67	74	28	3
③	Percentage(%)	42.67	22.33	24.67	9.33	1
④	Frequency	60	57	137	34	12
④	Percentage(%)	20	19	45.67	11.33	4

3.2.2

Statistical analysis of satisfaction with resource design

This paper investigates the satisfaction of resource design of resource recommendation platform. Respectively from ①“the resource structure of the resource library system is reasonably designed, the capacity is reasonably set”, ②“the resource design content can provide complete knowledge blocks, content organization systematic and logical”, ③“the resource design is both independent and interconnected, expanding the sense of use” and ④“resource design content knowledge division is complete, focus” four aspects of the investigation. The four aspects of satisfaction are the same as in 3.2.1. The results of the resource design satisfaction survey are shown in Table 2. The results show that: ① the distribution of the aspect samples is mainly “conformity”, “basic conformity”, “very conformity”, the proportion of which is 30%, 28.67% and 25% respectively, and the frequency is 90, 86 and 75 respectively. The distribution of the samples in aspect ② is dominated by the “basic conformity”, with a frequency and percentage of 120 and 40%, respectively. The distribution of the samples in aspect ③ is dominated by “very much in line with”, “basically in line with” and “in line with”, with frequencies and percentages of 95, 82, 81 and 31.67%, respectively, 27.33% and 27% respectively. The distribution of ④ samples is dominated by “basically complies”, with a frequency and percentage of 109 and 36.33% respectively. To summarize, the respondents’ satisfaction with the resource design is high, and the design is reasonable.

Table 2.

The results of the survey of resource design satisfaction

Evaluation content	Level of satisfaction	A	B	C	D	E
①	Frequency	75	90	86	30	19
①	Percentage(%)	25	30	28.67	10	6.33
②	Frequency	76	83	120	16	5
②	Percentage(%)	25.33	27.67	40	5.33	1.67
③	Frequency	95	81	82	34	8
③	Percentage(%)	31.67	27	27.33	11.33	2.67
④	Frequency	76	78	109	32	5
④	Percentage(%)	25.33	26	36.33	10.67	1.67

3.2.3

Statistical analysis of satisfaction with learning outcomes

This paper investigates the satisfaction of the learning effect of the resource recommendation platform. Five aspects are investigated: ①“Learning resources can basically meet the learning requirements of the course ‘Innovation and Entrepreneurship Education and Practice”. ②“Can recommend personalized learning content for students”. ③“Can effectively segment the learning content and facilitate students’ fragmented learning”. ④“can be helpful for the learning of the course ‘Innovation and Entrepreneurship Education and Practice”. ⑤“facilitates mobile learning after class and reflects the effect of personalized mobile learning”. The ground aspect satisfaction division is the same as 3.2.1. The results of the survey on satisfaction with learning outcomes are shown in Table 3. The results show that: ① the distribution of the aspect samples is dominated by “basically meets,” with a frequency and percentage of 110 and 36.67%, respectively, indicating that the learning resources in the Resource Bank system can basically meet the learning.② The distribution of aspect samples is dominated by “basic compliance”, with a percentage and frequency of 31.67% and 95 respectively, indicating that the resource library system is able to recommend personalized learning content for students. ③ The distribution of aspect samples is dominated by “basic compliance”, with a frequency and percentage of 43.67% and 131, respectively, indicating that the repository system can facilitate students’ fragmented learning. ④ The distribution of aspect samples is dominated by “basically conform” and “conform”, accounting for 32% and 28% respectively, and their corresponding frequencies are 96 and 84 respectively, indicating that the Resource Library System can be helpful for learning the course of “Innovative Entrepreneurship Education and Practice”. ⑤ The distribution of the samples is dominated by “basic compliance”, with a frequency and percentage of 34.67% and 104 respectively, indicating that the resource library system can facilitate students’ mobile learning after class. In summary, the respondents’ satisfaction with the learning effect is highly evaluated, and they make full use of the fragmented time to achieve the effect of personalized mobile learning after class.

Table 3.

Results of the study effect satisfaction survey

Evaluation content	Level of satisfaction	A	B	C	D	E
①	Frequency	64	79	110	36	11
①	Percentage(%)	21.33	26.33	36.67	12	3.67
②	Frequency	71	76	95	46	12
②	Percentage(%)	23.67	25.33	31.67	15.33	4
③	Frequency	60	65	131	37	7
③	Percentage(%)	20	21.67	43.67	12.33	2.33
④	Frequency	71	84	96	46	3
④	Percentage(%)	23.67	28	32	15.33	1
⑤	Frequency	74	78	104	43	1
⑤	Percentage(%)	24.67	26	34.67	14.33	0.33

3.2.4

Overall Satisfaction Results and Analysis

In the descriptive statistical analysis of the overall satisfaction of personalized learning system in this paper, the satisfaction is classified into five grades: “A very satisfied, B quite satisfied, C average, D dissatisfied, E very dissatisfied”, and then the number of satisfied students and the effective percentage of students are investigated. The overall satisfaction results of the personalized recommendation platform are shown in Figure 8. Through the frequency analysis, it can be seen that the distribution of the sample in terms of satisfaction is dominated by “very satisfied” with a frequency of 128, accounting for 42.67%, and “satisfied” with a frequency of 92, accounting for 30.67%. It shows that more than 70% of the students feel good about the personalized learning system for innovation and entrepreneurship education courses.

4

Conclusion

In this paper, a platform for a resource library system is constructed to promote innovation and entrepreneurship in schools, and the similarity of resources on the platform is calculated. After that, a personalized recommendation algorithm is used to assess the ability and utility of students, and on the basis of which resources are suitable for their innovation and entrepreneurship, they are recommended to them. Finally, the effectiveness of the model and satisfaction after application are evaluated. The primary conclusions are as follows: 1)

The simulation results show that when the number of labels given to the students is 6, the initial matching degree will grow rapidly to about 90%, and with the increase of time, the matching degree between personalized recommended resources and students will be stable at 95%. The recommended resources under 6 label numbers have a higher degree of matching with students. The accuracy of the recommendation algorithm in this paper is 0.9839, and the difference in the accuracy coefficient is extremely small (between 0.8942 and 0.9183) when the recommended resources are between 2 and 6.

2)

The mean values of personalized recommendation platforms’ satisfaction with the 4 aspects of interface design under the levels of very compliant, compliant, basic compliant, non-compliant, and very non-compliant are 36.42%, 22.33%, 25.67%, 11.92%, and 3.67%, respectively. The mean satisfaction levels of the personalized recommendation platform for the 4 parts of resource design at different levels were 26.83%, 27.67%, 33.08%, 9.33%, and 3.09%, respectively. The mean values of the personalized recommendation platform’s satisfaction with the 5 components of student learning outcomes at were 26.67%, 25.47%, 35.74%, 13.86%, and 2.27%, respectively, corresponding to frequency means ranging from 60-74, 65-84, 95-110, 36 -46 and 1-12. The comprehensive survey found that students’ overall satisfaction with personalized recommendation platforms is mainly “very satisfied” (42.67%). Apparently, the surveyed students received high ratings in terms of satisfaction with the interface design, resource design, and learning outcomes, and the platform of the repository system was well-designed.

Funding:

This research was supported by the Xi’an Jiaotong Engineering Institute 2022 university-level first-class undergraduate course construction Project: “Innovation and Entrepreneurship Education for College Students”.

Lingua:: Inglese

Frequenza di pubblicazione:: 1 volte all'anno
Argomenti della rivista:: Scienze biologiche, Scienze della vita, altro, Matematica, Matematica applicata, Matematica generale, Fisica, Fisica, altro

Feed RSS della rivista

A study of the role of Internet technology in promoting innovative and entrepreneurial activities in school education

Ru Qu

Pubblicato online: 19 mar 2025

Ricevuto: 08 nov 2024

Accettato: 13 feb 2025

DOI: https://doi.org/10.2478/amns-2025-0372

Parole chiaveInternet technology, Personalized resource recommendation, Similarity calculation, Innovation and entrepreneurship

© 2025 Ru Qu, published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Parole chiave
Internet technology, Personalized resource recommendation, Similarity calculation, Innovation and entrepreneurship