Reverse engineering variability from requirement documents based on probabilistic relevance and word embedding.
Yang LiSandro SchulzeGunter SaakePublished in: SPLC (2018)
Keyphrases
- reverse engineering
- sparck jones
- information retrieval
- relevance feedback
- word frequencies
- retrieved documents
- software engineering
- word spotting
- related documents
- information retrieval systems
- inverse document frequency
- software maintenance
- keywords
- vector space
- text documents
- document collections
- dynamic analysis
- latent topics
- term frequency
- ranked list
- document relevance
- program understanding
- text corpus
- relevance ranking
- object oriented
- probabilistic model
- term weighting
- conceptual schema
- software evolution
- multiword
- relevant documents
- word pairs
- sentence level
- relevance assessments
- legacy systems
- document retrieval
- document type
- co occurrence
- test collection
- gene regulatory networks
- document clustering
- reverse engineer
- retrieval systems
- database
- n gram
- software product line
- vector space model
- relevance model
- tf idf
- retrieval model
- bayesian networks
- software engineers