Solr-Plant

Solr-Plant performs fuzzy identification and taxonomic name resolution of plant species mentions in text using Apache Solr and sequence-alignment refinement to produce standardized taxonomic names for downstream analyses.

Key Features:

• Apache Solr-based fuzzy matching: Utilizes Apache Solr fuzzy search to identify plant name mentions within textual datasets.
• Smith-Waterman alignment refinement: Refines fuzzy matches using the Smith-Waterman alignment algorithm to improve matching precision.
• Taxonomic name resolution: Maps identified names to accepted taxonomic counterparts and synonyms by referencing a plant name and synonym thesaurus.
• Handling of misspellings: Demonstrates capability to detect and resolve misspelled species names within text.
• Benchmark evaluation: Evaluated on SALVIAS, the NCBI Taxonomy, and an S800 text corpus with reported high accuracy, precision, and recall.
• Scalability: Implements a scalable matching approach based on Apache Solr for processing large text corpora.

Scientific Applications:

• Plant taxonomic name resolution: Provides standardized, accepted plant names and synonym resolution for taxonomic curation and integration.
• Text-mining of biological corpora: Extracts accurate plant-related mentions from large text corpora to support data mining and downstream analyses.
• Species-level identification and extension: Optimized for species-level identification with methodology reported as extendable to broader taxonomic groups.

Methodology:

Employs Apache Solr fuzzy matching followed by Smith-Waterman alignment refinement, maps matches to accepted names and synonyms using a plant name and synonym thesaurus, and is evaluated on SALVIAS, NCBI Taxonomy, and S800 datasets.