Patent Portfolio

This chapter examines four complementary perspectives on how firms manage their patent portfolios: competitive proximity mapping reveals which firms occupy similar technology space, Shannon entropy tracks portfolio diversification over time, stacked CPC distributions illustrate how the largest filers distribute innovation across technology domains, and Jensen-Shannon divergence detects strategic pivots in portfolio composition. Together, these analyses provide a comprehensive view of corporate technology positioning and portfolio evolution.

The degree of similarity among corporate patent portfolios can be quantified by computing cosine similarityA measure of similarity between two vectors based on the angle between them. Values range from 0 (completely different) to 1 (identical). Used to compare patent portfolio compositions between companies. between CPC subclass distributions and projecting the results via UMAPUniform Manifold Approximation and Projection — a dimensionality reduction technique that preserves both local and global structure, used to visualize high-dimensional data in 2D., yielding a two-dimensional representation of the competitive landscape of innovation.

The degree of portfolio diversification among major filers varies substantially. Shannon entropyA measure of diversity or uncertainty in a distribution. Higher entropy means a more evenly spread portfolio across technology classes; lower entropy means concentration in fewer areas. across CPC subclasses measures whether a company distributes its innovation across many technology areas or concentrates in a few domains. Higher entropy indicates a broader, more diversified portfolio.

The technology portfolios of major patent holders illustrate how firms diversify their innovation across fields. The stacked bar chart below shows the distribution of patent grants across CPC technology sections for the ten largest assignees, revealing the breadth or concentration of each firm's portfolio.

Technology pivots occur when a company's patent portfolio shifts substantially between consecutive time windows. Using Jensen-Shannon divergenceA symmetric measure of the difference between two probability distributions. Used here to detect technology portfolio pivots by comparing a company's CPC distribution across consecutive time windows. (JSD) to measure the distance between CPC distributions across windows, it is possible to detect and characterize these pivots, which can precede business strategy announcements.

While CPC-based entropy captures granular subclass-level diversity, the WIPO technology classification offers a complementary perspective organized around 35 technology fields. WIPO-level Shannon entropy measures how evenly a firm distributes its patents across these broader technology domains, distinguishing diversified conglomerates from narrowly focused specialists.

Portfolio analysis reveals that firms pursue fundamentally different diversification strategies, and that portfolio shifts detected through Jensen-Shannon divergence often anticipate major business transformations. The next chapter, Interactive Company Profiles, combines patent output, technology portfolios, citation impact, and innovation strategy into comprehensive firm-level dashboards.