Science Explorer Interactive view Map

Topic Modeling

Topic Modeling is a research topic within Artificial Intelligence. Science Explorer counts 73k research works in it since 1956. 27.0% of them reached the world's top 10% most cited for their field and year.

This cluster of papers covers a wide range of advancements in natural language processing, including neural network architectures, word representation models, machine translation techniques, text classification algorithms, semantic similarity measures, named entity recognition methods, pretrained language models, sequence-to-sequence learning approaches, topic modeling strategies, and information retrieval systems.

  • Neural Networks
  • Word Representation
  • Machine Translation
  • Text Classification
  • Semantic Similarity
  • Named Entity Recognition
  • Pretrained Models
  • Sequence-to-Sequence Learning
  • Topic Modeling
  • Information Retrieval
Research works
73k
fractional, since 1956
In the world top 10%
20k
per year above
Top-10% rate
27.0%
share of its works in the world top 10%
Growth, 2013–17 → 2018–22
+103%
the tick is no change

Which countries lead Topic Modeling research?

By volume, China and the United States publish the most (8.4k and 4.2k works in 2022–2025).

By volume, 2022–2025

  1. 1 China 8.4k works
  2. 2 United States 4.2k works
  3. 3 India 2.2k works
  4. 4 Germany 933 works
  5. 5 United Kingdom 862 works
  6. 6 Japan 731 works
  7. 7 South Korea 543 works
  8. 8 Canada 533 works
  9. 9 Italy 459 works
  10. 10 France 455 works

How concentrated that is

The same countries as shares of everything the list above accounts for. A node where two countries do two thirds of the work and one spread evenly across twelve read alike as a ranking and not at all alike here.

China: 43.5%United States: 21.7%India: 11.3%Germany: 4.8%6 others listed: 18.6%43%largest
China8,371 · 43.5%United States4,184 · 21.7%India2,186 · 11.3%Germany933 · 4.8%6 others listed3,583 · 18.6%

Shares of the rows listed above, not of the whole node.

Which institutions lead Topic Modeling research?

By volume in 2022–2025, Tsinghua University publishes the most Topic Modeling research, followed by Beijing University of Posts and Telecommunications and Peking University.

Who are the leading researchers in Topic Modeling?

The most-cited researchers publishing on Topic Modeling include Geoffrey E. Hinton, Ilya Sutskever and Yoshua Bengio.

  1. 1 Geoffrey E. Hinton Canada 22k citations
  2. 2 Ilya Sutskever United States 21k citations
  3. 3 Yoshua Bengio Canada 17k citations
  4. 4 Wei Liu China 9.7k citations
  5. 5 Wei Liu Australia 9.7k citations
  6. 6 Ion Stoica United States 9.6k citations
  7. 7 Zhiheng Huang Germany 9k citations
  8. 8 Quoc V. Le United States 8.6k citations

Ranked by citations received across their whole record, among researchers with at least three works on this topic.

Where is Topic Modeling research done?

The largest centres of Topic Modeling research in 2022–2025 are Beijing (China), Shanghai (China), Guangzhou (China) and Wuhan (China). Among places with at least 20 works in it, it is an unusually large share of all research in Redmond, San Mateo and Nomi Shi.

Largest cities, 2022–2025

  1. 1 Beijing China 2.1k works
  2. 2 Shanghai China 664 works
  3. 3 Guangzhou China 383 works
  4. 4 Wuhan China 363 works
  5. 5 Hangzhou China 362 works
  6. 6 Nanjing China 345 works
  7. 7 Tokyo Japan 328 works
  8. 8 Seoul South Korea 319 works
  9. 9 Shenzhen China 314 works
  10. 10 Singapore Singapore 295 works

Where it is the local speciality

  1. RedmondUS · 50.0 works21×
  2. San MateoUS · 25.0 works18×
  3. Nomi ShiJP · 26.7 works15×
← less than its size predictsmore →

Location quotient: how much more of its research is in Topic Modeling than the world average.

See Topic Modeling on the map

Where is the best place to study Topic Modeling?

Among universities, judged by research, Renmin University of China, Singapore Management University and Carnegie Mellon University score highest, combining excellence, specialisation, size, growth and international reach. Research strength is one signal when choosing where to study; it does not measure teaching.

0%20%40%60%mean 34.8%fractional works in this node (log) →share in the world top 10% →Renmin University of China: 84, 34.5%Singapore Management University: 43, 41.3%Carnegie Mellon University: 115, 35.0%Nanyang Technological University: 89, 45.3%Beijing University of Posts and Telecommunications: 163, 14.9%Singapore University of Technology and Design: 22, 31.8%Indian Institute of Technology Patna: 32, 30.0%University of Hong Kong: 58, 51.3%Hong Kong University of Science and Technology: 66, 35.3%Mohamed bin Zayed University of Artificial Intelligence: 26, 28.6%Nanyang Technologica…Singapore Management…Carnegie Mellon Univ…Renmin University of…
above the meannear itbelow it

One dot per university in the table below. The upper left is the interesting corner: small places doing unusually strong work.

#UniversityScoreTop 10%SpecialisationWorksGrowth
1 Renmin University of ChinaChina 76.234.5%10.3×84 +249.5%
2 Singapore Management UniversitySingapore 74.741.3%16.1×43 +120.8%
3 Carnegie Mellon UniversityUnited States 70.635.0%11.4×115 +49.9%
4 Nanyang Technological UniversitySingapore 70.645.3%4.4×89 +147.5%
5 Beijing University of Posts and TelecommunicationsChina 68.414.9%10.6×163 +302.7%
6 Singapore University of Technology and DesignSingapore 68.231.8%9.4×22 +457.7%
7 Indian Institute of Technology PatnaIndia 64.830.0%9.8×32 +228.5%
8 University of Hong KongHong Kong 64.051.3%2.7×58 +94.5%
9 Hong Kong University of Science and TechnologyHong Kong 62.835.3%6.8×66 +33.8%
10 Mohamed bin Zayed University of Artificial IntelligenceUnited Arab Emirates 62.628.6%25.9×26

Universities only. Score blends excellence (30%), specialisation (25%), size (20%), growth (15%) and international reach (10%), 2015–2022; growth compares 2010–14 with 2015–19.

Is Topic Modeling research growing?

Output in 2018–2022 was 103% higher than in 2013–2017, peaking in 2025. The fastest-growing topics are Topic Modeling.

19801990200020102020
grewheldshrank

The same series as a ribbon — one cell per year, darker for more. The line above answers how much; this answers when.

Which topics inside it are moving

Growth and decline on one axis around a shared zero. Two lists side by side hide the thing that matters: whether the growth dwarfs the decline, or the other way round.