The established framework enables the construction of continuous pediatric percentile charts for laboratory analytes by applying a data-mining approach to clinical databases.
In our work, however, we study the process of opinion dynamics in online collaboration systems, by applying a data-driven model as well as by simulating how opinions spread in those systems.
Our aim is therefore to generalize the approach presented by Robinson et al. [11] for applications where an existing DEM is updated with a new DEM within a smaller subregion with irregularly shaped boundaries, and to improve the preservation of terrain shape when fusing DEMs by applying a data-driven, spatially variable blending zone width.
Although there are many efforts trying to minimize the impact of the variations among claim databases from different data holders by applying a common data model, bias can still be introduced since different data holders may choose different rules to convert their databases into a CDM.
As the selection of the external population is arbitrary, we evaluated the variability in our estimates by means of different approaches, namely, by applying: a) Dutch data-based proportions for the whole period; and b) Spanish proportions for the entire period (Additional file 1: Table S1).
From this study, we can conclude first that the classification of uterine EMG signals can be improved by applying a decision-level data fusion rule.
We began by applying a general purpose data mining procedure (Random Forest) to find a list of important variables that could plausibly affect the chosen dependant variables and black box models that relate these variables to the four BOPH.
By applying a custom-made data analysis pipeline, we have annotated the detected variants, including reported single-nucleotide polymorphisms (SNPs), genomic location, predicted miRNA binding sites, consequences of the variant in transcripts (i.e. synonymous, missense) and protein function prediction for those variants that are predicted to result in an aminoacid sustitution.
We accomplished the first step of identifying peaks in the frequency histogram by applying a novel random data-shuffling method (see Supplementary Material for details).
