Model interpretation is one of the key aspects of the model evaluation process. The explanation of the relationship between model variables and outputs is relatively easy for statistical models, such as linear regressions, thanks to the availability of model parameters and their statistical significance. For "black box" models, such as random forest, this information is hidden inside the model structure. This work presents an approach for computing feature contributions for random forest classification models. It allows for the determination of the influence of each variable on the model prediction for an individual instance. By analysing feature contributions for a training dataset, the most significant variables can be determined and their typical contribution towards predictions made for individual classes, i.e., class-specific feature contribution "patterns", are discovered. These patterns represent a standard behaviour of the model and allow for an additional assessment of the model reliability for a new data. Interpretation of feature contributions for two UCI benchmark datasets shows the potential of the proposed methodology. The robustness of results is demonstrated through an extensive analysis of feature contributions calculated for a large number of generated random forest models. * a.m.wojak@bradford.ac.uk

The ability to interpret the predictions made by quantitative structure activity relationships (QSARs) offers a number of advantages. Whilst QSARs built using non6linear modelling approaches, such as the popular Random Forest algorithm, might sometimes be more predictive than those built using linear modelling approaches, their predictions have been perceived as difficult to interpret. However, a growing number of approaches have been proposed for interpreting non6linear QSAR models in general and Random Forest in particular. In the current work, we compare the performance of Random Forest to two widely used linear modelling approaches: linear Support Vector Machines (SVM), or Support Vector Regression (SVR), and Partial Least Squares (PLS). We compare their performance in terms of their predictivity as well as the chemical interpretability of the predictions, using novel scoring schemes for assessing Heat Map images of substructural contributions. We critically assess different approaches to interpreting Random Forest models as well as for obtaining predictions from the forest. We assess the models on a large number of widely employed, public domain benchmark datasets corresponding to regression and binary classification problems of relevance to hit identification and toxicology. We conclude that Random Forest typically yields comparable or possibly better predictive performance than the linear modelling approaches and that its predictions may also be interpreted in a chemically and biologically meaningful way. In contrast to earlier work looking at interpreting non6linear QSAR models, we directly compare two methodologically distinct approaches for interpreting Random Forest models. The approaches for interpreting Random

Model interpretation is one of the key aspects of the model evaluation process. The explanation of the relationship between model variables and outputs is easy for statistical models, such as linear regressions, thanks to the availability of model parameters and their statistical significance. For "black box" models, such as random forest, this information is hidden inside the model structure. This work presents an approach for computing feature contributions for random forest classification models. It allows for the determination of the influence of each variable on the model prediction for an individual instance. Interpretation of feature contributions for two UCI benchmark datasets shows the potential of the proposed methodology. The robustness of results is demonstrated through an extensive analysis of feature contributions calculated for a large number of generated random forest models.

Sequential movement pattern-mining (SMP) in field-based team-sport: A framework for quantifying spatiotemporal data and improve training specificity?. SportRxiv, doi:xxx.

This study investigated sources of variability in the overall and phase-specific running match characteristics in elite rugby league. Microtechnology data were collected from 11 Super League (SL) teams, across 322 competitive matches within the 2018 and 2019 seasons. Total distance, high-speed running (HSR) distance (>5•5 m•s -1 ), average speed, and average acceleration were assessed. Variability was determined using linear mixed models, with random intercepts specified for player, position, match, and club. Large within-player coefficients of variation (CV) were found across whole match, ball-in-play, attack and defence for total distance (CV range = 24% to 35%) and HSR distance (37% to 96%), whereas small to moderate CVs (≤10%) were found for average speed and average acceleration. Similarly, there was higher between-player, -position, and -match variability in total distance and HSR distance when compared with average speed and average acceleration across all periods. All metrics were stable between-teams (≤5%), except HSR distance (16% to 18%). The transition period displayed the largest variability of all phases, especially for distance (up to 42%) and HSR distance (up to 165%). Absolute measures of displacement display large within-player and between-player, -position, and -match variability, yet average acceleration and average speed remain relatively stable across all match-periods.