Lin Gong

Ontology-Based collaborative filtering recommendation algorithm

E-learning system for knowledge points recommended primarily uses traditional collaborative filtering algorithm. Similarity calculation of knowledge points is often based on user rating above the intersection of knowledge points. The different semantic relations between knowledge points are not well considered, which results in the low recommended accuracy. This paper proposed an Ontology-based collaborative filtering recommendation algorithm, which could help users find the nearest neighbors even if the scores of knowledge points are little or zero. Through experiment, this algorithm was compared to traditional collaborative filtering recommendation algorithms. The new method achieved a better recommendation.

A quantitative approach to design alternative evaluation based on data-driven performance prediction

A systematic approach is proposed to quantitatively evaluate design alternatives.The judgments of designers are quantified in the vague and subjective environment.Obtains the optimal alternative based on the data-driven performance prediction.Strengthens efficiency and objectivity by reducing vagueness and human involvement. Design alternative evaluation in the early stages of engineering design plays an important role in determining the success of new product development, as it influences considerably the subsequent design activities. However, existing approaches to design alternative evaluation are overly reliant on experts ambiguous and subjective judgments and qualitative descriptions. To reduce subjectivity and improve efficiency of the evaluation process, this paper proposes a quantitative evaluation approach through data-driven performance predictions. In this approach, the weights of performance characteristics are determined based on quantitative assessment of expert judgments, and the ranking of design alternatives is achieved by predicting performance values based on historical product design data. The experts subjective and often vague judgments are captured quantitatively through a rough number based Decision-Making Trial and Evaluation Laboratory (DEMATEL) method. In order to facilitate performance based quantitative ranking of alternatives at the early stages of design where no performance calculation is possible, a particle swarm optimization based support vector machine (PSO-SVM) is applied for historical data based performance prediction. The final ranking of alternatives given the predicted values of multiple performance characteristics is achieved through Viekriterijumska Optimizacija I kompromisno Reenje (VIKOR). A case study is carried out to demonstrate the validity of the proposed approach.

Innovation Oriented Hyper Model and Evolution of Knowledge Network

Aiming at the topological characteristics and unclear innovation of knowledge network, a hyper model for knowledge was established. Based on the model, the keyword network and its evolution model were constructed with a newly-raised concept called innovation degree which measures the innovation in specific knowledge field. After that, specific model was constructed and its evolution was analyzed through papers from the field of artificial intelligence, which verified that the topological structure presented small-world nature and the innovation degree were effective in identifying the direction of knowledge innovation in relevant fields.

Modeling and Analysis of Conceptual Design-Oriented Product Technology Network

Innovation in conceptual design period of products is an important indicator to show innovation capacity of one enterprise, and the improvement of relative technology is the base of it. By analyzing technical papers in a period, the relevant keywords can be summarized to make up a relevant technical keywords network, which may be very complex and include a large number of nodes and edges. This keywords network will show the key topics and hotspots in science and technology area currently or in the future. By building the keywords network which base on product conceptual design and analyzing the statistical result, some available techniques can be found to support product conceptual design. In this report an example of coal cutter is provided to discuss the feasibility and usability of this model.

The effects of team cognitive on creative design: Improving the creativity and quality of product

Nowadays, the competition among modern manufacturing enterprises is actually relied on quality and efficiency of innovative design based on teamwork. The role of innovative design based on teamwork manufacturing sector is more important than ever. In the modern manufacturing context, in order to study teamwork innovative design quality and efficiency evaluation technical issues based on cognitive process, the human brain memory model of cognitive processes is studied. Based on analyzing the memory model of cognitive design process, creative design process cognitive models and innovative design team cognitive process model, the creative design cognitive processes in the individual and team level are elaborated. After analyzing the connotation of quality and efficiency, the creative design of the five quality dimensions and five rating scale and the relationship between team innovation design quality and the consistent of individuals design result are extracted. Then theory hypotheses are verified with experiment raw data. Finally experimental results obtained are as follows, in the aspect of design quality, the higher the design consistency among individual members design results are, the higher the design quality is.

Research on Individuation Task Allocating Based on Team Collaboration

Team collaboration for project implementation has become the main pattern in modern work. How to allocate tasks to the right team member is the key for project completion. The paper firstly reviewed the related methods and concepts of team collaboration and knowledge sharing. Then the model of knowledge was built. Based on knowledge model, this paper built the model of the task and the team member. According to the knowledge background and project experience of team members, the paper proposed an allocation model which automatically allocated tasks in group work. Finally, a prototype system was developed for practical validation.

Research on the Integration Method for Heterogeneous Database Based on Ontology

With the establishment of more and more management information systems in the enterprises, many heterogeneous databases have been produced, resulting in the phenomenon of the information island. This paper puts forward a method based on ontology to integrate the heterogeneous databases and presents the framework of the integration method. In this framework A method based on relational database is presented to build the mixed ontology structure Common VocabularyLocal Ontology. The mixed ontology structure is used to establish the semantic web for enterprises. This novel approach exploits the multithreading combined with semantic web to implement parallel retrieval on an integrated platform. The application that this prototype system of the integration method has been implemented in a research institute verifies the method could effectively analyze the users intention, thereby improving the reuse of knowledge.