Estimation and evaluation of Students’ behaviors in E- learning Environment using adaptive computing

Companies and research institutes are acquiring monopolized technologies through patents to survive in a rapidly changing market. The applicant is guaranteed by law the exclusive right to the technology for a certain period. A prior art search is an essential task for this purpose. Therefore, the need for a successful prior art search is increasing. Previous studies have disadvantages because relevant fields are limited, requiring an in-depth understanding of specific technologies. This paper proposes a method to overcome these limitations using patent text and citation information. The proposed method searches for prior art using a citation network based on the shortest path derived from a breadth-first search. We examined based on experiments whether the proposed method could be applied to the industry. Based on the results, the proposed algorithm can efficiently search for prior art. Furthermore, it was possible to construct a counterstrategy based on the characteristics of the applicant. The code will be made available at github.com/lee-ju/k_step_PIC.

A Fast and Scalable Algorithm for Prior Art Search

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.

/pdf/a-study-on-the-calibrated-confidence-of-text-classification-1rqj3kdl.pdf

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

Companies and research institutes are acquiring monopolized technologies through patents to survive in a rapidly changing market. The applicant is guaranteed by law the exclusive right to the technology for a certain period. A prior art search is an essential task for this purpose. Therefore, the need for a successful prior art search is increasing. Previous studies have disadvantages because relevant fields are limited, requiring an in-depth understanding of specific technologies. This paper proposes a method to overcome these limitations using patent text and citation information. The proposed method searches for prior art using a citation network based on the shortest path derived from a breadth-first search. We examined based on experiments whether the proposed method could be applied to the industry. Based on the results, the proposed algorithm can efficiently search for prior art. Furthermore, it was possible to construct a counterstrategy based on the characteristics of the applicant. The code will be made available at github.com/lee-ju/k&#x005F;step&#x005F;PIC. 

/pdf/a-fast-and-scalable-algorithm-for-prior-art-search-kyf6sv23.pdf

In contemporary times, science-based technologies are needed for launching innovative products and services in the market. As technology-based management strategies are gaining importance, associated patents need to be comprehensively studied. Previous studies have proposed predictive models based on patent factors. However, technology-based management strategies can influence the growth and decline of firms. Thus, this study aims to estimate uncertainties of the factors that are frequently used in technology-based studies. Furthermore, the importance of the factors may fluctuate over time. Therefore, we propose a Bayesian neural network model based on Flipout and four research hypotheses to evaluate the validity of our method. The proposed method not only estimates the uncertainties of the factors, but also predicts the future value of technologies. Our contribution is to (i) provide a tractable Bayesian neural network applicable to big data, (ii) discover factors that affect the value of technology, and (iii) present empirical evidence for the timeliness and objectivity of technology evaluation. In our experiments, 3781 healthcare-related cases of patents were used, and we found that the proposed hypotheses were all statistically significant. Therefore, we believe that reliable and stable technology-based management strategies can be established through our method.

/pdf/estimation-of-uncertainty-for-technology-evaluation-factors-dovt0xco.pdf

Estimation of Uncertainty for Technology Evaluation Factors via Bayesian Neural Networks

This paper proposes a multi-class classification model for technology evaluation (TE) using patent documents. TE is defined as converting technology quality to its present value; it supports efficient research and development using intellectual property rights–research & development (IP–R&D) and decision-making by companies. Through IP–R&D, companies create their patent portfolios and develop technology management strategies. They protect core patents and use those patents to cooperate with other companies. In modern society, as conversion technology has been rapidly developed, previous TE methods became difficult to apply to technology. This is because they relied on expert-based qualitative methods. Qualitative results are difficult to use to guarantee objectivity. Many previous studies have proposed models for evaluating technology based on patent data to address these limitations. However, those models can lose contextual information during the preprocessing of bibliographic information and require a lexical analyzer suitable for processing terminology in patents. This study uses a lexical analyzer produced using a deep learning structure to overcome this limitation. Furthermore, the proposed method uses quantitative information and bibliographic information of patents as explanatory variables and classifies the technology into multiple classes. The multi-class classification is conducted by sequentially evaluating the value of a technology. This method returns multiple classes in order, enabling class comparison. Moreover, it is model-agnostic, enabling diverse algorithms to be used. We conducted experiments using actual patent data to examine the practical applicability of the proposed methodology. Based on the experiment results, the proposed method was able to classify actual patents into an ordered multi-class. In addition, it was possible to guarantee the objectivity of the results. This is because our model used the information in the patent specification. Furthermore, the model using both quantitative and bibliographic information exhibited higher classification performance than the model using only quantitative information. Therefore, the proposed model can contribute to the sustainable growth of companies by classifying the value of technology into more detailed categories.

https://www.mdpi.com/2071-1050/12/15/6153/pdf

A Multi-Class Classification Model for Technology Evaluation

A patent system requires novelty and progressiveness so that new patents do not infringe on the rights of prior art. Patent investigation including a prior art search is essential to the process of commercialization of technology. In general, patent investigation has been conducted by experts based on their qualitative judgement. However, the number of patents has increased so fast that it has become difficult to handle the quantitative burdens of the search with a conventional approach. There have been previous studies dealing with patent investigation to find similar technologies. They had limitations as they did not utilize the citation relationship and similarity between patents in a comprehensive way. In addition, they could not properly reflect the sequential citation relationship of patents though this is effective in discovering similar patents. In this study, we propose an efficient methodology to discover similar technologies by comprehensively considering the similarity and citation relationship between patents. In particular, we intended to reflect the citation sequence and indirect citation relationship in the process of searching for similar patents. For this, we introduced the concept of “patents with indirect connections” (PICs) and devised an algorithm to efficiently detect patent pairs having such a relationship. The proposed methodology of this study contributes to preventing patent litigation in advance by discovering patents with such potential risks. It is expected that this method will provide patent applicants with the opportunity to establish appropriate strategies against competitors with similar technologies. In order to examine the practical applicability of the proposed method, Korean patents related to machine learning and deep learning were collected. As a result of the experiment, it was possible to identify 24 pairs of similar patents without a direct citation relationship and derive appropriate counter strategies.

https://www.mdpi.com/2071-1050/13/2/820/pdf

Juhyun Lee

Papers

A Multi-Class Classification Model for Technology Evaluation

Introducing Patents with Indirect Connection (PIC) for Establishing Patent Strategies

A Fast and Scalable Algorithm for Prior Art Search

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

Multimodal Deep Learning for Patent Classification