Topic Modeling (LDA / NMF)
Probabilistic or matrix-factorisation topics over the full-text corpus.
- Family
- Computational thematic
- Engine
- Lacuna native (Python)
When to use it
Use it to map themes, trace how they evolve, and mark candidate gaps with textual evidence.
Surfaces conceptual proximity, repetition and zones of silence across the corpus text. It is computational — not Braun–Clarke reflexive thematic analysis.
Figures it produces
- Dendrogram
- Sankey flow
Limitations and scope
- Computational analysis, not Braun–Clarke reflexive thematic analysis — it surfaces patterns in text, it does not interpret meaning for you.
- Results are conditional on the studies, corpus and parameters you supply, and on the chosen model. The analysis summarises the evidence as reported; it does not establish causation.
- Every figure carries its source, and every run is reproducible bit-for-bit from the same inputs, parameters and engine version.
In this family
- Co-word Analysis & Thematic MapKeyword co-occurrence clustered into a strategic diagram (centrality × density).
- Embedding-based Topic ClusteringDocuments embedded (TF-IDF by default; transformer/SBERT optional) and clustered into semantically coherent topics.
- Factorial AnalysisMCA / correspondence analysis projecting terms into a conceptual map.
- Thematic EvolutionHow themes split, merge and flow across time slices (Sankey).
- Trending Topics & Burst DetectionEmerging terms and citation bursts over the publication timeline.
Run this analysis
The definition above is open. The live engine, its parameters and the provenance-sealed report run inside your workspace.