MKEM: a Multi-level Knowledge Emergence Model for mining undiscovered public knowledge

Abstract

BackgroundSince Swanson proposed the Undiscovered Public Knowledge (UPK) model, there have been many approaches to uncover UPK by mining the biomedical literature. These earlier works, however, required substantial manual intervention to reduce the number of possible connections and are mainly applied to disease-effect relation. With the advancement in biomedical science, it has become imperative to extract and combine information from multiple disjoint researches, studies and articles to infer new hypotheses and expand knowledge.MethodsWe propose MKEM, a Multi-level Knowledge Emergence Model, to discover implicit relationships using Natural Language Processing techniques such as Link Grammar and Ontologies such as Unified Medical Language System (UMLS) MetaMap. The contribution of MKEM is as follows: First, we propose a flexible knowledge emergence model to extract implicit relationships across different levels such as molecular level for gene and protein and Phenomic level for disease and treatment. Second, we employ MetaMap for tagging biological concepts. Third, we provide an empirical and systematic approach to discover novel relationships.ResultsWe applied our system on 5000 abstracts downloaded from PubMed database. We performed the performance evaluation as a gold standard is not yet available. Our system performed with a good precision and recall and we generated 24 hypotheses.ConclusionsOur experiments show that MKEM is a powerful tool to discover hidden relationships residing in extracted entities that were represented by our Substance-Effect-Process-Disease-Body Part (SEPDB) model.

A systematic review on literature-based discovery workflow

“…Examples of measures that represent this category include Degree centrality (Goodwin, Cohen & Rindflesch, 2012), Eigenvector centrality (Özgür et al, 2010), Closeness centrality (Özgür et al, 2011), Betweenness centrality (Özgür et al, 2010), Common Neighbours (Kastrin, Rindflesch & Hristovski, 2014), Jaccard Index (Kastrin, Rindflesch & Hristovski, 2014), Preferential Attachment (Kastrin, Rindflesch & Hristovski, 2014), Personalised PageRank (Petric et al, 2014), Personalised Diffusion Ranking (Petric et al, 2014), and Spreading Activation (Goodwin, Cohen & Rindflesch, 2012). Knowledge-based Measures: This category denotes the scoring measures such as MeSHbased Literature cohesiveness (Swanson, Smalheiser & Torvik, 2006), semantic type cooccurrence (Jha & Jin, 2016b), chemDB atomic count (Ijaz, Song & Lee, 2010), and chemDB XLogP (Ijaz, Song & Lee, 2010) that involve the knowledge from structured resources to rank the associations. The advantage of these measures is that they entangle semantic aspects into consideration to decide the potentiality of the association.…”

“…Song, Heo & Ding (2015) have also proposed a combined ranking measure by considering an average of three semantic similarity measures, and SemRep score. The characteristics that have been considered in the study of Ijaz, Song & Lee (2010) include UMLS semantic type, structural similarity, chemDB atomic count, and chemDB XLogP. Similarly, Gopalakrishnan et al (2017) have also introduced a combined ranking measure by integrating global (node centrality and MeSH tree code depth) and local (semantic co-occurrence and betweenness centrality) measures.…”

“…Other text analytics resources include NLTK : to identify Noun Phrases (Sebastian, Siew & Orimaye, 2017b) and stop words (Lever et al, 2017), ReVerb: to extract relations (Preiss, Stevenson & Gaizauskas, 2015), Stanford parser: for dependency tree parsing (Sang et al, 2015) and extract relations (Preiss, Stevenson & Gaizauskas, 2015), Stanford CoreNLP: for sentence boundary detection, POS tagging and lemmatisation (Song, Heo & Ding, 2015), WordNet: for word sense disambiguation (Sebastian, Siew & Orimaye, 2017b), RacerPro: for logical and rule-based reasoning (Guo & Kraines, 2009), Link Grammar Parser: for sentence parsing (Ijaz, Song & Lee, 2010), Vantage Point: for document clustering, auto-correction mapping and factor matrix analysis (Kostoff, 2011), Nvivo: to extract terms, stop words, coding and matrix coding queries (Kibwami & Tutesigensi, 2014), CLUTO: for document clustering (Kostoff et al, 2008b), Lucene: for information retrieval (Malec et al, 2016), and OntoGen: for topic ontology construction (Petriȇ et al, 2009).…”

Generation and application of drug indication inference models using typed network motif comparison analysis

“…A few previous studies proposed specific information models to store these various types of data. Ijaz, et al [14] defined a specific frame of entities as an information model. In this way, they could store only the targeted information.…”

BMC Ecology embraces biodiversity research

“…A new discipline, biodiversity informatics, has a goal to develop tools and pipelines that let the scientist see the bigger picture emerging from these various types of data [3]. For this approach to be successful, all data has to be accessible to the research community without restrictions.…”