Artificial Intelligence for Modeling Complex Systems: Taming the Complexity of Expert Models to Improve Decision Making

Major societal and environmental challenges involve complex systems that have diverse multi-scale interacting processes. Consider, for example, how droughts and water reserves affect crop production and how agriculture and industrial needs affect water quality and availability. Preventive measures, such as delaying planting dates and adopting new agricultural practices in response to changing weather patterns, can reduce the damage caused by natural processes. Understanding how these natural and human processes affect one another allows forecasting the effects of undesirable situations and study interventions to take preventive measures. For many of these processes, there are expert models that incorporate state-of-the-art theories and knowledge to quantify a system's response to a diversity of conditions. A major challenge for efficient modeling is the diversity of modeling approaches across disciplines and the wide variety of data sources available only in formats that require complex conversions. Using expert models for particular problems requires integration of models with third-party data as well as integration of models across disciplines. Modelers face significant heterogeneity that requires resolving semantic, spatiotemporal, and execution mismatches, which are largely done by hand today and may take more than 2 years of effort. We are developing a modeling framework that uses artificial intelligence (AI) techniques to reduce modeling effort while ensuring utility for decision making. Our work to date makes several innovative contributions: (1) an intelligent user interface that guides analysts to frame their modeling problem and assists them by suggesting relevant choices and automating steps along the way; (2) semantic metadata for models, including their modeling variables and constraints, that ensures model relevance and proper use for a given decision-making problem; and (3) semantic representations of datasets in terms of modeling variables that enable automated data selection and data transformations. This framework is implemented in the MINT (Model INTegration) framework, and currently includes data and models to analyze the interactions between natural and human systems involving climate, water availability, agricultural production, and markets. Our work to date demonstrates the utility of AI techniques to accelerate modeling to support decision-making and uncovers several challenging directions for future work.

Analysis of sweet corn nutritional values using multivariate statistical methods

Processing large amounts of data provided by automated analytical equipment requires carefulness. Most mathematical and statistical methods have strict application conditions. Most of these methods are based on eigenvalue calculations and require variables to be correlated in groups. If this condition is not met, the most popular multivariate methods cannot be used. The best procedure for such testing is the Kaiser-Meyer-Olkin test for Sampling Adequacy. Two databases were examined using the KMO test. One of them resulted from the sweet corn measured in the scone of the study, while the other from the 1979 book of János Sváb. For both databases, MSA (measures sampling adequacy) was well below the critical value, thus they are not suitable e.g. for principal component analysis. In both databases, the values of the partial correlation coefficients were much higher than Pearson’s correlation coefficients. Often the signs of partial coefficients did not match the signs of linear correlation coefficients. One of the main reasons for this is that the correlation between the variables is non-linear. Another reason is that control variables have a non-linear effect on a given variable. In such cases, classical methods should be disregarded and expert models better suited to the problem should be chosen in order to analyse the correlation system.

Bayesian Crime Investigations: Integrating Actuarial And Expert Models

In response to concerns that some of the most methodologically rigorous predictive studies of criminal offender characteristics may yet be less generalizable and applicable than advertised or assumed, this research first tests how well seven regression analysis models (represented by 28 equations) predict characteristics across three conditions: familiar cases (used to create the regressions), less familiar cases (native to the sample used to create the regressions) and foreign cases (from a similar but novel sample). Here a linear trend shows overfitting of the models to their own sample: a drop-off in prediction accuracy relative to simple mean-based prediction as cases become more foreign (ηp 2 = .646). In response to hopes that subjective input from expert police investigators could be integrated into the models to correct for this overfitting bias, this research also tests an algorithm combining expert ratings with the regression equations. Here moderate and significant improvement in novel-case prediction is observed overall (p = .036, r = .44) and equations for all twelve expert participants are shown to improve prediction to varying degrees. These results suggest that current best methods would perform poorly in the field, but can be improved by expert insight.

Bayesian Crime Investigations: Integrating Actuarial And Expert Models

In response to concerns that some of the most methodologically rigorous predictive studies of criminal offender characteristics may yet be less generalizable and applicable than advertised or assumed, this research first tests how well seven regression analysis models (represented by 28 equations) predict characteristics across three conditions: familiar cases (used to create the regressions), less familiar cases (native to the sample used to create the regressions) and foreign cases (from a similar but novel sample). Here a linear trend shows overfitting of the models to their own sample: a drop-off in prediction accuracy relative to simple mean-based prediction as cases become more foreign (ηp 2 = .646). In response to hopes that subjective input from expert police investigators could be integrated into the models to correct for this overfitting bias, this research also tests an algorithm combining expert ratings with the regression equations. Here moderate and significant improvement in novel-case prediction is observed overall (p = .036, r = .44) and equations for all twelve expert participants are shown to improve prediction to varying degrees. These results suggest that current best methods would perform poorly in the field, but can be improved by expert insight.

Conflict Resolution in Mechatronic Collaborative Design Using Category Theory

Due to the multitude of disciplines involved in mechatronic design, heterogeneous languages and expert models are used to describe the system from different domain-specific views. Despite their heterogeneity, these models are highly interrelated. As a consequence, conflicts among expert models are likely to occur. In order to ensure that these models are not contradictory, the necessity to detect and manage conflicts among the models arises. Detecting these inconsistencies at an early stage significantly reduces the amount of engineering activities re-execution. Therefore, to deal with this issue, a formal framework relying upon mathematical concepts is required. The mathematical theory, namely category theory (CT), is considered as an efficient tool to provide a formal and unifying framework supporting conflict detection and management. This paper proposes a comprehensive methodology that allows conflict detection and resolution in the context of mechatronic collaborative design. CT is used in order to explicitly capture the inconsistencies occurred between the disparate expert models. By means of this theory, the conflicts can be detected and handled in an easy and formal way. Our proposed approach is applied to a collaborative scenario concerning the electro-mechanical actuator (EMA) of the aileron.

Evolving Musical Sight Reading Exercises Using Expert Models

Sight reading skills are widely considered to be crucial for all musicians. However, given that sight reading involves playing sheet music without having seen it before, once an exercise has been completed by a student it can no longer be used as a sight reading exercise for them. In this paper we present a novel evolutionary algorithm for generating musical sight reading exercises in the Western art music tradition. Using models based on expert examples, the algorithm generates material suitable for practice which is both technically appropriate and aesthetically pleasing with respect to an instrument and difficulty level. This overcomes the resource constraint in using traditional practice exercises, which are exhausted quickly by students and teachers due to their limited quantity.