Automated learning of domain taxonomies from text using background knowledge.
Ontology highlight
ABSTRACT:
SUBMITTER: Hoxha J
PROVIDER: S-EPMC5077645 | biostudies-literature | 2016 Oct
REPOSITORIES: biostudies-literature
Publications
Automated learning of domain taxonomies from text using background knowledge.
Hoxha Julia J Jiang Guoqian G Weng Chunhua C
Journal of biomedical informatics 20160903
In this paper, we present an automated method for taxonomy learning, focusing on concept formation and hierarchical relation learning. To infer such relations, we partition the extracted concepts and group them into closely-related clusters using Hierarchical Agglomerative Clustering, informed by syntactic matching and semantic relatedness functions. We introduce a novel, unsupervised method for cluster detection based on automated dendrogram pruning, which is dynamic to each partition. We evalu ...[more]
Similar Datasets
Project description:BackgroundCurrent text mining tools supporting abstract screening in systematic reviews are not widely used, in part because they lack sensitivity and precision. We set out to develop an accessible, semi-automated "workflow" to conduct abstract screening for systematic reviews and other knowledge synthesis methods.MethodsWe adopt widely recommended text-mining and machine-learning methods to (1) process title-abstracts into numerical training data; and (2) train a classification model to predict eligible abstracts. The predicted abstracts are screened by human reviewers for ("true") eligibility, and the newly eligible abstracts are used to identify similar abstracts, using near-neighbor methods, which are also screened. These abstracts, as well as their eligibility results, are used to update the classification model, and the above steps are iterated until no new eligible abstracts are identified. The workflow was implemented in R and evaluated using a systematic review of insulin formulations for type-1 diabetes (14,314 abstracts) and a scoping review of knowledge-synthesis methods (17,200 abstracts). Workflow performance was evaluated against the recommended practice of screening abstracts by 2 reviewers, independently. Standard measures were examined: sensitivity (inclusion of all truly eligible abstracts), specificity (exclusion of all truly ineligible abstracts), precision (inclusion of all truly eligible abstracts among all abstracts screened as eligible), F1-score (harmonic average of sensitivity and precision), and accuracy (correctly predicted eligible or ineligible abstracts). Workload reduction was measured as the hours the workflow saved, given only a subset of abstracts needed human screening.ResultsWith respect to the systematic and scoping reviews respectively, the workflow attained 88%/89% sensitivity, 99%/99% specificity, 71%/72% precision, an F1-score of 79%/79%, 98%/97% accuracy, 63%/55% workload reduction, with 12%/11% fewer abstracts for full-text retrieval and screening, and 0%/1.5% missed studies in the completed reviews.ConclusionThe workflow was a sensitive, precise, and efficient alternative to the recommended practice of screening abstracts with 2 reviewers. All eligible studies were identified in the first case, while 6 studies (1.5%) were missed in the second that would likely not impact the review's conclusions. We have described the workflow in language accessible to reviewers with limited exposure to natural language processing and machine learning, and have made the code available to reviewers.
| S-EPMC8152711 | biostudies-literature
Project description:BackgroundToxicogenomics studies often profile gene expression from assays involving multiple doses and time points. The dose- and time-dependent pattern is of great importance to assess toxicity but computational approaches are lacking to effectively utilize this characteristic in toxicity assessment. Topic modeling is a text mining approach, but may be used analogously in toxicogenomics due to the similar data structures between text and gene dysregulation.ResultsTopic modeling was applied to a very large toxicogenomics dataset containing microarray gene expression data from >15,000 samples associated with 131 drugs tested in three different assay platforms (i.e., in vitro assay, in vivo repeated dose study and in vivo single dose experiment) with a design including multiple doses and time points. A set of "topics" which each consist of a set of genes was determined, by which the varying sensitivity of three assay systems was observed. We found that the drug-dependent effect was more pronounced in the two in vivo systems than the in vitro system, while the time-dependent effect was most strongly reflected in the in vitro system followed by the single dose study and lastly the repeated dose experiment. The dose-dependent effect was similar across three assay systems. Although the results indicated a challenge to extrapolate the in vitro results to the in vivo situation, we did notice that, for some drugs but not for all the drugs, the similarity in gene expression patterns was observed across all three assay systems, indicating a possibility of using in vitro systems with careful designs (such as the choice of dose and time point), to replace the in vivo testing strategy. Nonetheless, a potential to replace the repeated dose study by the single-dose short-term methodology was strongly implied.ConclusionsThe study demonstrated that text mining methodologies such as topic modeling provide an alternative method compared to traditional means for data reduction in toxicogenomics, enhancing researchers' capabilities to interpret biological information.
| S-EPMC4236689 | biostudies-literature
Project description:BackgroundVirtual screening in the form of similarity rankings is often applied in the early drug discovery process to rank and prioritize compounds from a database. This similarity ranking can be achieved with structural similarity measures. However, their general nature can lead to insufficient performance in some application cases. In this paper, we provide a link between ranking-based virtual screening and fragment-based data mining methods. The inclusion of binding-relevant background knowledge into a structural similarity measure improves the quality of the similarity rankings. This background knowledge in the form of binding relevant substructures can either be derived by hand selection or by automated fragment-based data mining methods.ResultsIn virtual screening experiments we show that our approach clearly improves enrichment factors with both applied variants of our approach: the extension of the structural similarity measure with background knowledge in the form of a hand-selected relevant substructure or the extension of the similarity measure with background knowledge derived with data mining methods.ConclusionOur study shows that adding binding relevant background knowledge can lead to significantly improved similarity rankings in virtual screening and that even basic data mining approaches can lead to competitive results making hand-selection of the background knowledge less crucial. This is especially important in drug discovery and development projects where no receptor structure is available or more frequently no verified binding mode is known and mostly ligand based approaches can be applied to generate hit compounds.
| S-EPMC3928642 | biostudies-literature
Project description:BackgroundSegmentation of computed tomography (CT) is important for many clinical procedures including personalized cardiac ablation for the management of cardiac arrhythmias. While segmentation can be automated by machine learning (ML), it is limited by the need for large, labeled training data that may be difficult to obtain. We set out to combine ML of cardiac CT with domain knowledge, which reduces the need for large training datasets by encoding cardiac geometry, which we then tested in independent datasets and in a prospective study of atrial fibrillation (AF) ablation.MethodsWe mathematically represented atrial anatomy with simple geometric shapes and derived a model to parse cardiac structures in a small set of N = 6 digital hearts. The model, termed "virtual dissection," was used to train ML to segment cardiac CT in N = 20 patients, then tested in independent datasets and in a prospective study.ResultsIn independent test cohorts (N = 160) from 2 Institutions with different CT scanners, atrial structures were accurately segmented with Dice scores of 96.7% in internal (IQR: 95.3%-97.7%) and 93.5% in external (IQR: 91.9%-94.7%) test data, with good agreement with experts (r = 0.99; p < 0.0001). In a prospective study of 42 patients at ablation, this approach reduced segmentation time by 85% (2.3 ± 0.8 vs. 15.0 ± 6.9 min, p < 0.0001), yet provided similar Dice scores to experts (93.9% (IQR: 93.0%-94.6%) vs. 94.4% (IQR: 92.8%-95.7%), p = NS).ConclusionsEncoding cardiac geometry using mathematical models greatly accelerated training of ML to segment CT, reducing the need for large training sets while retaining accuracy in independent test data. Combining ML with domain knowledge may have broad applications.
| S-EPMC10577270 | biostudies-literature
Project description:Documented occurrences of fossil taxa are the empirical foundation for understanding large-scale biodiversity changes and evolutionary dynamics in deep time. The fossil record contains vast amounts of understudied taxa. Yet the compilation of huge volumes of data remains a labour-intensive impediment to a more complete understanding of Earth's biodiversity history. Even so, many occurrence records of species and genera in these taxa can be uncovered in the palaeontological literature. Here, we extract observations of fossils and their inferred ages from unstructured text in books and scientific articles using machine-learning approaches. We use Bryozoa, a group of marine invertebrates with a rich fossil record, as a case study. Building on recent advances in computational linguistics, we develop a pipeline to recognize taxonomic names and geologic time intervals in published literature and use supervised learning to machine-read whether the species in question occurred in a given age interval. Intermediate machine error rates appear comparable to human error rates in a simple trial, and resulting genus richness curves capture the main features of published fossil diversity studies of bryozoans. We believe our automated pipeline, that greatly reduced the time required to compile our dataset, can help others compile similar data for other taxa.
| S-EPMC6501925 | biostudies-literature
Project description:BACKGROUND: The construction of a whole-genome physical map has been an essential component of numerous genome projects initiated since the inception of the Human Genome Project. Its usefulness has been proved for whole-genome shotgun projects as a post-assembly validation and recently it has also been used in the assembly step to constrain on BACs positions. Fingerprinting is usually the method of choice for construction of physical maps. A clone fingerprint is composed of true peaks representing real fragments and background peaks, mainly composed of E. coli genomic DNA, partial digestions, star activity by-products, and machine background. High-throughput fingerprinting leads to the production of thousands of BAC clone fingerprints per day. That is why background peaks removal has become an important issue and needs to be automatized, especially in capillary electrophoresis based fingerprints. RESULTS: At the moment, the only tools available for such a task are GenoProfiler and its descendant FPMiner. The large variation in the quality of fingerprints that is usually present in large fingerprinting projects represents a major difficulty in the correct removal of background peaks that has only been partially addressed by the methods so far adopted that all require a long manual optimization of parameters. Thus, we implemented a new data-independent tool, FPB (FingerPrint Background removal), suitable for large scale projects as well as mapping of few clones. CONCLUSION: FPB is freely available at http://www.appliedgenomics.org/tools.php. FPB was used to remove the background from all fingerprints of three grapevine physical map projects. The first project consists of about 50,000 fingerprints, the second one consists of about 70,000 fingerprints, and the third one consists of about 45,000 fingerprints. In all cases a successful assembly was built.
| S-EPMC2689866 | biostudies-other
Project description:ObjectiveSimplifying healthcare text to improve understanding is difficult but critical to improve health literacy. Unfortunately, few tools exist that have been shown objectively to improve text and understanding. We developed an online editor that integrates simplification algorithms that suggest concrete simplifications, all of which have been shown individually to affect text difficulty.Materials and methodsThe editor was used by a health educator at a local community health center to simplify 4 texts. A controlled experiment was conducted with community center members to measure perceived and actual difficulty of the original and simplified texts. Perceived difficulty was measured using a Likert scale; actual difficulty with multiple-choice questions and with free recall of information evaluated by the educator and 2 sets of automated metrics.ResultsThe results show that perceived difficulty improved with simplification. Several multiple-choice questions, measuring actual difficulty, were answered more correctly with the simplified text. Free recall of information showed no improvement based on the educator evaluation but was better for simplified texts when measured with automated metrics. Two follow-up analyses showed that self-reported education level and the amount of English spoken at home positively correlated with question accuracy for original texts and the effect disappears with simplified text.DiscussionSimplifying text is difficult and the results are subtle. However, using a variety of different metrics helps quantify the effects of changes.ConclusionText simplification can be supported by algorithmic tools. Without requiring tool training or linguistic knowledge, our simplification editor helped simplify healthcare related texts.
| S-EPMC9155254 | biostudies-literature
Project description:Measuring the concentration and viability of fungal cells is an important and fundamental procedure in scientific research and industrial fermentation. In consideration of the drawbacks of manual cell counting, large quantities of fungal cells require methods that provide easy, objective and reproducible high-throughput calculations, especially for samples in complicated backgrounds. To answer this challenge, we explored and developed an easy-to-use fungal cell counting pipeline that combined the machine learning-based ilastik tool with the freeware ImageJ, as well as a conventional photomicroscope. Briefly, learning from labels provided by the user, ilastik performs segmentation and classification automatically in batch processing mode and thus discriminates fungal cells from complex backgrounds. The files processed through ilastik can be recognized by ImageJ, which can compute the numeric results with the macro 'Fungal Cell Counter'. Taking the yeast Cryptococccus deneoformans and the filamentous fungus Pestalotiopsis microspora as examples, we observed that the customizable software algorithm reduced inter-operator errors significantly and achieved accurate and objective results, while manual counting with a haemocytometer exhibited some errors between repeats and required more time. In summary, a convenient, rapid, reproducible and extremely low-cost method to count yeast cells and fungal spores is described here, which can be applied to multiple kinds of eucaryotic microorganisms in genetics, cell biology and industrial fermentation.
| S-EPMC8576076 | biostudies-literature
Project description:ObjectivesTo automatically construct a drug indication taxonomy from drug labels using generative Artificial Intelligence (AI) represented by the Large Language Model (LLM) GPT-4 and real-world evidence (RWE).Materials and methodsWe extracted indication terms from 46 421 free-text drug labels using GPT-4, iteratively and recursively generated indication concepts and inferred indication concept-to-concept and concept-to-term subsumption relations by integrating GPT-4 with RWE, and created a drug indication taxonomy. Quantitative and qualitative evaluations involving domain experts were performed for cardiovascular (CVD), Endocrine, and Genitourinary system diseases.Results2909 drug indication terms were extracted and assigned into 24 high-level indication categories (ie, initially generated concepts), each of which was expanded into a sub-taxonomy. For example, the CVD sub-taxonomy contains 242 concepts, spanning a depth of 11, with 170 being leaf nodes. It collectively covers a total of 234 indication terms associated with 189 distinct drugs. The accuracies of GPT-4 on determining the drug indication hierarchy exceeded 0.7 with "good to very good" inter-rater reliability. However, the accuracies of the concept-to-term subsumption relation checking varied greatly, with "fair to moderate" reliability.Discussion and conclusionWe successfully used generative AI and RWE to create a taxonomy, with drug indications adequately consistent with domain expert expectations. We show that LLMs are good at deriving their own concept hierarchies but still fall short in determining the subsumption relations between concepts and terms in unregulated language from free-text drug labels, which is the same hard task for human experts.
| S-EPMC11339527 | biostudies-literature
Project description:Despite the potential of deep learning (DL)-based methods in substituting CT-based PET attenuation and scatter correction for CT-free PET imaging, a critical bottleneck is their limited capability in handling large heterogeneity of tracers and scanners of PET imaging. This study employs a simple way to integrate domain knowledge in DL for CT-free PET imaging. In contrast to conventional direct DL methods, we simplify the complex problem by a domain decomposition so that the learning of anatomy-dependent attenuation correction can be achieved robustly in a low-frequency domain while the original anatomy-independent high-frequency texture can be preserved during the processing. Even with the training from one tracer on one scanner, the effectiveness and robustness of our proposed approach are confirmed in tests of various external imaging tracers on different scanners. The robust, generalizable, and transparent DL development may enhance the potential of clinical translation.
| S-EPMC9537165 | biostudies-literature