library(tidyverse)
library(plotly)

# Cleaned data (made in data_prep.R)
arxiv_stats    <- read_csv("data_clean/arxiv_stats.csv")
pat_year_type  <- read_csv("data_clean/patents_by_year_type.csv")
pat_year_juris <- read_csv("data_clean/patents_by_year_jurisdiction.csv")
pat_applicants <- read_csv("data_clean/patents_by_year_applicant.csv")

# Patents per year (applications + granted only)
pat_total <- pat_year_type |>
  filter(doc_type %in% c("Granted Patent", "Patent Application")) |>
  group_by(year) |>
  summarise(patents = sum(n))

# Papers per year (articles + cross-lists)
papers_total <- arxiv_stats |>
  mutate(papers = articles + cross_lists) |>
  select(year, papers)

# Papers and patents in one table
both <- left_join(papers_total, pat_total, by = "year")

Introduction

This story compares how quantum physics research publications and quantum computing patent records have changed between 2000 and 2025. I wanted to see whether patent activity follows a similar trend to academic research or whether the gap between them changes over time. This is interesting because it can help us understand how research and technological development are progressing in the quantum field.

Growth in Quantum Physics Research Publications

The gap between articles and cross listed papers grows slowly for many years but gets much bigger in recent years, especially from the early 2020s, as articles increase faster than cross listed papers.

This is interesting because it shows that both are growing but articles are increasing faster in recent years.

p1 <- arxiv_stats |>
  pivot_longer(c(articles, cross_lists), names_to = "type", values_to = "count") |>
  ggplot(aes(year, count, colour = type)) +
  geom_line(linewidth = 1) +
  scale_colour_viridis_d(end = 0.8) +
  theme_minimal() +
  labs(title = "Growth in Quantum Physics Research Publications",
       x = "Year",
       y = "Number of entries",
       colour = "Publication type",
       caption = "Source: arXiv (2026)")

ggplotly(p1)

Patent Applications vs Granted Patents

Patent applications and granted patents stay close to each other in the earlier years but over the last decade the gap gets bigger as applications increase much faster than granted patents.

This is interesting because more patents are being applied for but the number of granted patents is not increasing at the same rate. Part of this is because patents take years to be granted, so many recent applications are still waiting.

p2 <- pat_year_type |>
  filter(doc_type %in% c("Granted Patent", "Patent Application")) |>
  ggplot(aes(year, n, colour = doc_type)) +
  geom_line(linewidth = 1) +
  scale_colour_viridis_d(end = 0.8) +
  theme_minimal() +
  labs(title = "Patent Applications vs Granted Patents",
       x = "Year",
       y = "Number of patent records",
       colour = "Document type",
       caption = "Source: The Lens (2026)")

ggplotly(p2)

Comparing Research and Patent Growth Since 2010

Research papers and patent records both increase over time but from the mid-2010s patent growth increases much faster while research growth remains more gradual.

This is interesting because it shows that patent activity is growing much faster than research publications compared to their levels in 2010. Patent numbers were very small in 2010, which makes their growth look extra large.

p3 <- both |>
  mutate(papers_idx  = papers  / papers[year == 2010]  * 100,
         patents_idx = patents / patents[year == 2010] * 100) |>
  select(year, papers_idx, patents_idx) |>
  pivot_longer(-year, names_to = "series", values_to = "index") |>
  ggplot(aes(year, index, colour = series)) +
  geom_line(linewidth = 1) +
  geom_hline(yintercept = 100, linetype = "dashed") +
  scale_colour_viridis_d(end = 0.8) +
  theme_minimal() +
  labs(title = "Research and Patent Growth Since 2010",
       x = "Year",
       y = "Growth index (2010 = 100)",
       colour = "Measure",
       caption = "Source: arXiv (2026); The Lens (2026)")

ggplotly(p3)

Conclusion

The three graphs show that research publications and patent activity have increased over time but their growth patterns start to differ after around 2017. Patent applications pull further away from granted patents while patent records grow much faster than research publications relative to their respective 2010 levels. This is interesting because it shows a growing difference between the trends in academic research and patenting activity although it does not prove a direct connection between individual research papers and patents.

About the data

Sources

  1. arXiv: I used the official arXiv yearly statistics pages for the quantum physics category from 2000 to 2025 and collected the annual numbers of articles and cross listed papers. This data shows how research publication activity in the category has changed over time.
  2. Lens.org: I used patent data from Lens.org with the CPC classification G06N 10/00 for quantum computing and selected records from 2000 to 2025. This allowed me to examine changes in patent applications and granted patents over time.

What I counted

  1. Papers: Papers were counted by adding articles and cross listed papers for each year. Articles are papers counted under the category while cross listed papers are papers listed in the category but primarily associated with another arXiv category.
  2. Patents: Patents were counted by adding patent applications and granted patents for each year. Other document types were excluded so that the analysis focused on these two categories of patent records.

Limitations

  1. Patent publication lag: Recent patent counts may appear lower because some patent records take time to become publicly available. Granted patents may also appear later than their original applications because the approval process takes time.
  2. The same invention counted in several countries: The same invention can be patented in multiple countries and may appear as several patent records in the dataset. Therefore the number of patent records does not necessarily represent the number of unique inventions.
  3. Not every paper is on arXiv: The arXiv dataset does not include every quantum physics or quantum computing research paper. Some research is published through journals and other platforms so the results represent activity recorded in the arXiv category rather than all research worldwide.

Checks

  1. Checking arXiv numbers: I used the official arXiv yearly statistics pages to extract article and cross list counts for 2000 to 2025 and tested the extraction using the 2025 page. I also ran a separate arXiv API query for 2020 as an additional check because the results may differ depending on the search and counting method used.

References

Anthropic. (2026). Claude (Opus 5.5) [Large language model]. https://claude.ai

arXiv. (n.d.). Quantum physics (quant-ph): Article statistics by year [Data set]. Cornell University. Retrieved October 27, 2026, from https://arxiv.org/year/quant-ph/2025

R Core Team. (2026). R: A language and environment for statistical computing (Version 4.5.3) [Computer software]. R Foundation for Statistical Computing. https://www.R-project.org/

Sievert, C. (2020). Interactive web-based data visualization with R, plotly, and shiny. Chapman and Hall/CRC. https://plotly-r.com

The Lens. (2026). Patent search results: CPC classification G06N10* [Data set]. Cambia. Retrieved October 27, 2026, from https://www.lens.org/lens/search/patent/list?q=class_cpc.symbol:G06N10*

Wickham, H., Averick, M., Bryan, J., Chang, W., McGowan, L. D., François, R., Grolemund, G., Hayes, A., Henry, L., Hester, J., Kuhn, M., Pedersen, T. L., Miller, E., Bache, S. M., Müller, K., Ooms, J., Robinson, D., Seidel, D. P., Spinu, V., … Yutani, H. (2019). Welcome to the tidyverse. Journal of Open Source Software, 4(43), 1686. https://doi.org/10.21105/joss.01686

Acknowledgements

I used Claude (Opus 5.5), an AI model made by Anthropic, for technical help in this assignment:

It helped me set up my RStudio project and fix package errors

It showed me how to get data from the arXiv website and how to read the Lens.org patent file into R

It helped me with R code syntax, debugging errors

It suggested references in APA style

It gave feedback on the spelling and wording of my title and description ```