Chung-Huei Kuan

Ranking patent assignee performance by h-index and shape descriptors

We propose a geometric interpretation to the ranking of patent assignees by their h-indices as indicating the relative positions of their rank-citation curves. We then propose two shape descriptors characterizing the rank-citation curves over the h-cores and h-tails, respectively. Together with the h-indices, the shape descriptors help verifying the geometric relationship among rank-citation curves and the relative performance among the assignees’h-cores and h-tails. The geometric interpretation and shape descriptors are proven by empirical data to be reliable, accurate, robust, flexible, and insightful, and their application could be extended to research performance evaluation as well.

Positioning research and innovation performance using shape centroids of h-core and h-tail

We propose a novel yet practical method capturing an individuals research or innovation performance by the shape centroids of the h-core and h-tail areas of its publications or patents. A large number of individuals’ relative performance with respect to their h-cores and h-tails can be simultaneously positioned and conveniently observed in two-dimensional coordinate systems. Two approaches are further proposed to the utilization of the two-dimensional distribution of shape centroids. The first approach specifically determines, within a group of individuals, those outperforming or being outperformed by a target individual. The second approach provides a quick qualitative categorization of the individuals so that the nature of their performance is revealed. Using patent assignees as an illustrative case, the approaches are tested with empirical patent assignee data.

Cross-field evaluation of publications of research institutes using their contributions to the fields' MVPs determined by h-index

We propose a cross-field evaluation method for the publications of research institutes. With this approach, we first determine a set of the most visible publications (MVPs) for each field from the publications of all assessed institutes according to the fields h-index. Then, we measure an institutes production in each field by its percentage share (i.e., contribution) to the fields MVPs. Finally, we obtain an institutes cross-field production measure as the average of its contributions to all fields. The proposed approach is proven empirically to be reasonable, intuitive to understand, and uniformly applicable to various sets of institutes and fields of different publication and citation patterns. The field and cross-field production measures obtained by the proposed approach not only allow linear ranking of institutes, but also reveal the degree of their production difference.

A two-dimensional approach to performance evaluation for a large number of research institutions

We characterize the research performance of a large number of institutions in a two-dimensional coordinate system based on the shapes of their h-cores so that their relative performance can be conveniently observed and compared. The 2D distribution of these institutions is then utilized (1) to categorize the institutions into a number of qualitative groups revealing the nature of their performance, and (2) to determine the position of a specific institution among the set of institutions. The method is compared with some major h-type indices and tested with empirical data using clinical medicine as an illustrative case. The method is extensible to the research performance evaluation at other aggregation levels such as researchers, journals, departments, and nations.

Missing links: Timing characteristics and their implications for capturing contemporaneous technological developments

Abstract A missing link in this study refers to a pair of patents whose relatedness is not manifested by one citing the other but implied by their strong bibliographic coupling. By analyzing empirical data, this study discovers that the occurrence of missing links is not coincidental but arises systematically; patent pairs with missing links usually have highly overlapped application processes, whereas those with direct citations more frequently have successive or less overlapped application processes. The missing links thus may capture relatedness between patents that direct citations fail to detect. By applying main path analysis to a network containing 34,083 patents, 155,076 citations, and 9,213 missing links designed to simulate direct citations, this study further finds that the missing links—accounting for only approximately 5% of all connections—identify patents embodying contemporaneous technological developments, which may evade detection if only direct citations are considered.

Locating the key competitors: A new tool for technology manager

We propose to use the product of a competitors h-index and a weighted sum of the citations of its h-core patents as a Core Value characterizing the competitors h-core performance in terms of both quality and quantity. The key competitors are those of the greatest Core Values. The approach is applied to empirical data and compared with similar measures such as the products of the h-index and other h-type indices. The approach obviates the insensitivity problem to exceptionally highly cited patents and does not impose significant analytical overhead to the technology manager.