Meizeng Gui

Bio: Meizeng Gui is an academic researcher from Zhejiang University of Finance and Economics. The author has co-authored 1 publications.

Applying data mining techniques for technology prediction in new energy vehicle: a case study in China

Abstract: Technology prediction is an important technique to help new energy vehicle (NEV) firms keep market advantage and sustainable development. Under fierce competition in the new energy industry, there is an urgent necessity for innovative technology prediction method to effectively identify core and frontier technologies for NEV firms. Among the various methods of technology prediction, one of the most frequently used methods is to make technology prediction from patent data. This paper synthesizes the frequent pattern growth (FP-growth) algorithm and input-output analysis to construct a new technology prediction method based on the knowledge flow perspective, takes the data of NEV patent family in 1989–2018 the Derwent patent database as a sample, divides the data according to the 5-year standard, and uses the method to identify the core and frontier technologies in the NEV field during different periods. Furthermore, the multiple co-occurrence method applies to analyze the technology layout and evolution patterns in China’s NEV field. The results show that the technology prediction method proposed in this paper can effectively identify core and frontier technologies to achieve NEV technology prediction.

A Study on the Calibrated Confidence of Text Classification Using a Variational Bayes

Abstract: Recently, predictions based on big data have become more successful. In fact, research using images or text can make a long-imagined future come true. However, the data often contain a lot of noise, or the model does not account for the data, which increases uncertainty. Moreover, the gap between accuracy and likelihood is widening in modern predictive models. This gap may increase the uncertainty of predictions. In particular, applications such as self-driving cars and healthcare have problems that can be directly threatened by these uncertainties. Previous studies have proposed methods for reducing uncertainty in applications using images or signals. However, although studies that use natural language processing are being actively conducted, there remains insufficient discussion about uncertainty in text classification. Therefore, we propose a method that uses Variational Bayes to reduce the difference between accuracy and likelihood in text classification. This paper conducts an experiment using patent data in the field of technology management to confirm the proposed method’s practical applicability. As a result of the experiment, the calibrated confidence in the model was very small, from a minimum of 0.02 to a maximum of 0.04. Furthermore, through statistical tests, we proved that the proposed method within the significance level of 0.05 was more effective at calibrating the confidence than before.

Analysis of the dealer car service spare range using the FP-Growth algorithm

Abstract: In the current conditions of instability and a rapidly changing economy, mathematical methods and intelligent information technologies used in making managerial decisions in various fields play an important role. It is especially necessary to approach carefully the process of securing stocks of products sold, which is necessary for the profit of a car service company. The company in its activity requires a wide range of cars spare parts. The lack of necessary parts can provoke a long downtime of cars waiting for the technical maintenance or the customer's refusal from service. Excess parts that have not been sold for a long time require increased storage costs. In this article, the FP-Growth algorithm is used to analyze the range of cars spare parts for dealer car service company, which solves the problem of finding associative rules. This task is based on searching in a large volume of source data for relationships in the form of if X, then Y. The FP-Growth algorithm differs from other methods of searching for associative rules by the procedure of constructing a tree of variants of sets of objects, which allows to reduce the search for possible variations and reduce the number of iterations. To implement the proposed algorithm, the Loginom Community analytical system was used. As a result, sets of spare parts were identified, often used together in the current repair of cars.