[Objective] This paper aims to improve the quality and efficiency of concept design with the help of technology framework for semantic knowledge management based on knowledge flow in conceptual design. [Methods] First, we introduced knowledge flow theory to conceptual design based on design knowledge management research. Then, we modeled the knowledge flow in conceptual design from two perspectives. Finally, we established the technology framework for semantic knowledge management and analyzed its requirements, composition and functions. [Results] We built the knowledge flow reference model, the integration model of work and knowledge flows, as well as the semantic knowledge management framework. [Limitations] We built the technology framework at the macro level, which needs to be optimized at the micro level. [Conclusions] The proposed framework could systematically reveal the mechanism of modern design knowledge management, and indicate some future studies.
[Objective] This paper studies the unified modeling technology for multidimensional design knowledge in conceptual design, aiming to effectively manage and reuse knowledge. [Methods] First, we analyzed the needs of unified modeling for multidimensional design from the perspectives of knowledge management and reuse. Then, we classified the conceptual design knowledge based on design process and knowledge attributes. Third, we constructed the unified model system for multidimensional design knowledge with static design object and dynamic design flow at the core. Finally, we established the unified model with the help of Ontology model. [Results] We divided the conceptual design knowledge into six categories: design objects, design flow, design organization, design resources, design criteria and design cases. We proposed a unified modeling system based on design object and flow, and also created the knowledge Ontology to model the knowledge of design organization, design resources, design criteria and design cases. [Limitations] More research is needed to examine the proposed technology. [Conclusions] The unified modeling technology enables us to effectively model the dynamic conceptual design knowledge.
[Objective] This study explores the semantic modeling technology for conceptual design process, aiming to effectively manage and reuse dynamic knowledge. [Methods] First, we analyzed the demands of semantic modeling from the perspectives of knowledge management and reuse. Then, we integrated the template and component technology to build semantic modeling system. Finally, we semantically modeled the design flow, which was integrated to the knowledge flow model. [Results] We proposed a two-layer modular encapsulation technology combining configurable flow template and knowledge component, which was used to construct the semantic modeling systems for the design and knowledge flow. [Limitations] We need to add case studies to our research. [Conclusions] The constructed semantic modeling system could effectively model the dynamic design knowledge in the conceptual design process.
[Objective] This paper proposes a new knowledge acquisition method for conceptual design process, aiming to improve the modular modeling and representation, as well as effectively manage and reuse dynamic knowledge. [Methods] First, we analyzed the design process knowledge acquisition from the perspective of content. Then, we designed a knowledge acquisition experiment with the help of interviews and the double-layer flow acquisition template. Finally, we obtained the implicit design process knowledge in organization by interviews, which was modularized by the double-layer flow acquisition template. [Results] We used the conceptual design process of a recoil brake part as case to retrieve the initial design process knowledge. The design flow knowledge, which was organized and normalized by acquisition template of the standardized design flow, was dismantled and packaged by acquisition template of the design step. [Limitations] Both the experiment and data analysis were conducted manually, which posed negative effects to the efficiency. [Conclusions] The proposed method could effectively obtain the tacit and dynamic design process knowledge from conceptual design.
[Objective] This paper aims to analyze the impacts of de-centralization level on NBA team’s performance. [Methods] First, we watched the 2017 NBA Final videos to establish a ball-passing network. Then, we originally proposed the definition and algorithm of weighted directed network entropy(WDNE) to analyze the team’s performance. [Results] We found the WDNE value was positively correlated to the degree of network decentralization. When the team’s WDNE gap was large, the score gap was also widened, or vice versa. [Limitations] Manual recording might lead to omissions and deviations, while the sample size was small. [Conclusions] The decentralization level poses positive effects to a team’s performance, as well as the cohesion and robustness of the team’s network.
[Objective] This study analyzes the dissemination of marketing information on social network systems, aiming to identify the most influential nodes. [Methods] We collected Twitter data on Huawei Mate 9 smartphone to analyze users’ information behaviors like tweeting, retweeting and commenting. First, the network topology was described as topology structure diagram; Second, we examined scales of the network; Finally, we used independent cascade model (ICM) to simulate information dissemination. [Results] We found that initial active nodes selection based on the new measurements performed well. [Limitations] The parameters of ICM could be optimized. [Conclusions] The enterprises should pay attention to both official and accidental nodes to retrieve feedback from the market.
[Objective] This paper aims to improve the classification results of anonymous groups and then obtain better data masking model and algorithm. [Methods] First, we modified the dimension judgment standards based on k-anonymity. Then, we used the KD tree as storage structure to construct a new algorithm. Third, we implemented the proposed algorithm with Python. Finally, we examined the feasibility and effectiveness of the new algorithm with the number of anonymous groups and the percentage of NCP. [Results] The new algorithm could maximize the number of anonymous groups generated by the whole dataset, while the percentage of NCP was lower than similar algorithms. [Limitations] For datasets with significant degree of dispersion, the dimension of the loop computation was cumbersome. [Conclusions] The proposed algorithm could improve the availability of the anonymous groups and reduce the data loss.
[Objective] This paper aims to investigate the evolution of online public opinion by analyzing the spatial-temporal distribution patterns of topic evolution. [Methods] First, we used the LDA model to extract topics from news and then calculated the quantitative topic intensity index to measure their popularity. Second, we adopted spatial autocorrelation method to examine the distribution of topic intensity on “tourism” as well as its changes over time based on Moran’s I Index. [Results] The global distribution of topic intensity was clustered and characterized by the global Moran’s I index. The local distribution of topic intensity had hot spots, abnormal high values and low values. [Limitations] Only collected data from Xinhuanet, which might yield in-complete results. [Conclusions] The proposed method could effectively extract the spatial-temporal patterns of online public opinion, which improves the decision-making and early warning mechanism.
[Objective] This paper proposes a model to identify the interests of online shoppers based on their browsing behaviors, aiming to improve the personalized recommendation services. [Methods] First, we launched experiment to collect clickstream data from Taobao and TMall. Second, we used the Bisecting K-means algorithm to analyze the retrieved data. Finally, we established the relationship mapping structure between interests and behaviors. [Results] We found four types of user’s implicit interests: Attention, Comprehension, Attitudes and Intention. Users with the Attitude and Intention types tended to make purchase. The characteristics of browsing paths were different among the users. [Limitations] We did not examine unstructured data, i.e., online sales advertisements, in this study. [Conclusions] This paper investigates the user interests in online shopping, and then improve the personalized recommendation services of the E-commerce platforms.
[Objective] This paper aims to improve the performance of automatic abstracting with the help of “Doc2vec” model and improved K-means clustering algorithm. [Methods] First, we introduced the Doc2Vec model, which could examine the document contextual information, to extract the semantics, grammar and word sequences of Chinese document sentences. Then, we transformed these sentences to vectors of fixed dimensions. Third, we identified clustering centers for the improved K-means algorithm, and then processed the sentence vectors. Finally, the sentences with larger information entropy in one cluster, as well as higher similarity with other sentences in the cluster, were extracted. [Results] Compared with the PLSA method, the precision, recall, and F value of the proposed model increased by 9.57%, 7.62% and 10.30% respectively. [Limitations] We could not use the sentences extracted from the documents to generate high quality abstracts. [Conclusions] The proposed method could improve the performance of automatic abstracting of Chinese documents.
[Objective] This paper aims to improve the performance of Chinese word segmentation techniques on domain literature by optimizing results of existing approaches. [Methods] First, we proposed a new criteria of Term Frequency Deviation (TFD) to capture word formation characteristics of domain literature based on the analysis of segmentation errors. Then, we developed an unsupervised segmentation refining approach with the help of TFD. [Results] We examined the proposed approach with agriculture documents. It improved the segmentation results of three popular Chinese word segmentation approaches (i.e., ICTCLAS, THULAC and LTP) by 2%~3% in F1 measure. The proposed approach was easy to use and robustness to parameters. [Limitations] The recall of the proposed approach needs to be improved. [Conclusions] The new Chinese word segmentation approach, which imrpoves the performance of traditional methods on domain literature, could be applied to other fields due to its independence of domain-specific vocabulary and annotated corpus.