Towards cost-sensitive adaptation: when is it worth updating your predictive model?

Our digital universe is rapidly expanding,more and more daily activities are digitally recorded, data arrives in streams, it needs to be analyzed in real time and may evolve over time. In the last decade many adaptive learning algorithms and prediction systems, which can automatically update themselves with the new incoming data, have been developed. The majority of those algorithms focus on improving the predictive performance and assume that model update is always desired as soon as possible and as frequently as possible. In this study we consider potential model update as an investment decision, which, as in the financial markets, should be taken only if a certain return on investment is expected. We introduce and motivate a new research problem for data streams ? cost-sensitive adaptation. We propose a reference framework for analyzing adaptation strategies in terms of costs and benefits. Our framework allows to characterize and decompose the costs of model updates, and to asses and interpret the gains in performance due to model adaptation for a given learning algorithm on a given prediction task. Our proof-of-concept experiment demonstrates how the framework can aid in analyzing and managing adaptation decisions in the chemical industry.

Online dietary intake estimation : reproducibility and validity of the Food4Me food frequency questionnaire against a 4-day weighed food record

Background: Advances in nutritional assessment are continuing to embrace developments in computer technology. The online Food4Me food frequency questionnaire (FFQ) was created as an electronic system for the collection of nutrient intake data. To ensure its accuracy in assessing both nutrient and food group intake, further validation against data obtained using a reliable, but independent, instrument and assessment of its reproducibility are required. Objective: The aim was to assess the reproducibility and validity of the Food4Me FFQ against a 4-day weighed food record (WFR). Methods: Reproducibility of the Food4Me FFQ was assessed using test-retest methodology by asking participants to complete the FFQ on 2 occasions 4 weeks apart. To assess the validity of the Food4Me FFQ against the 4-day WFR, half the participants were also asked to complete a 4-day WFR 1 week after the first administration of the Food4Me FFQ. Level of agreement between nutrient and food group intakes estimated by the repeated Food4Me FFQ and the Food4Me FFQ and 4-day WFR were evaluated using Bland-Altman methodology and classification into quartiles of daily intake. Crude unadjusted correlation coefficients were also calculated for nutrient and food group intakes. Results: In total, 100 people participated in the assessment of reproducibility (mean age 32, SD 12 years), and 49 of these (mean age 27, SD 8 years) also took part in the assessment of validity. Crude unadjusted correlations for repeated Food4Me FFQ ranged from .65 (vitamin D) to .90 (alcohol). The mean cross-classification into “exact agreement plus adjacent” was 92% for both nutrient and food group intakes, and Bland-Altman plots showed good agreement for energy-adjusted macronutrient intakes. Agreement between the Food4Me FFQ and 4-day WFR varied, with crude unadjusted correlations ranging from .23 (vitamin D) to .65 (protein, % total energy) for nutrient intakes and .11 (soups, sauces and miscellaneous foods) to .73 (yogurts) for food group intake. The mean cross-classification into “exact agreement plus adjacent” was 80% and 78% for nutrient and food group intake, respectively. There were no significant differences between energy intakes estimated using the Food4Me FFQ and 4-day WFR, and Bland-Altman plots showed good agreement for both energy and energy-controlled nutrient intakes. Conclusions: The results demonstrate that the online Food4Me FFQ is reproducible for assessing nutrient and food group intake and has moderate agreement with the 4-day WFR for assessing energy and energy-adjusted nutrient intakes. The Food4Me FFQ is a suitable online tool for assessing dietary intake in healthy adults.

On the predictive content of leading indicators: the case of U.S. real estate markets

This paper employs a probit and a Markov switching model using information from the Conference Board Leading Indicator and other predictor variables to forecast the signs of future rental growth in four key U.S. commercial rent series. We find that both approaches have considerable power to predict changes in the direction of commercial rents up to two years ahead, exhibiting strong improvements over a naïve model, especially for the warehouse and apartment sectors. We find that while the Markov switching model appears to be more successful, it lags behind actual turnarounds in market outcomes whereas the probit is able to detect whether rental growth will be positive or negative several quarters ahead.

On the validity of single-parcel energetics to assess the importance of internal energy and compressibility effects in stratified fluids

It is often assumed on the basis of single-parcel energetics that compressible effects and conversions with internal energy are negligible whenever typical displacements of fluid parcels are small relative to the scale height of the fluid (defined as the ratio of the squared speed of sound over gravitational acceleration). This paper shows that the above approach is flawed, however, and that a correct assessment of compressible effects and internal energy conversions requires considering the energetics of at least two parcels, or more generally, of mass conserving parcel re-arrangements. As a consequence, it is shown that it is the adiabatic lapse rate and its derivative with respect to pressure, rather than the scale height, which controls the relative importance of compressible effects and internal energy conversions when considering the global energy budget of a stratied fluid. Only when mass conservation is properly accounted for is it possible to explain why available internal energy can account for up to 40 percent of the total available potential energy in the oceans. This is considerably larger than the prediction of single-parcel energetics, according to which this number should be no more than about 2 percent.

A psychometric analysis of reliability and validity of the index of learning styles (ILS)

Prior literature showed that Felder and Silverman learning styles model (FSLSM) was widely adopted to cater to individual styles of learners whether in traditional or Technology Enhanced Learning (TEL). In order to infer this model, the Index of Learning Styles (ILS) instrument was proposed. This research aims to analyse the soundness of this instrument in an Arabic sample. Data were integrated from different courses and years. A total of 259 engineering students participated voluntarily in the study. The reliability was analysed by applying internal construct reliability, inter-scale correlation, and total item correlation. The construct validity was also considered by running factor analysis. The overall results indicated that the reliability and validity of perception and input dimensions were moderately supported, whereas processing and understanding dimensions showed low internal-construct consistency and their items were weakly loaded in the associated constructs. Generally, the instrument needs further effort to improve its soundness. However, considering the consistency of the produced results of engineering students irrespective of cross-cultural differences, it can be adopted to diagnose learning styles.

Testing projected wild bee distributions in agricultural habitats: predictive power depends on species traits and habitat type

Species distribution models (SDM) are increasingly used to understand the factors that regulate variation in biodiversity patterns and to help plan conservation strategies. However, these models are rarely validated with independently collected data and it is unclear whether SDM performance is maintained across distinct habitats and for species with different functional traits. Highly mobile species, such as bees, can be particularly challenging to model. Here, we use independent sets of occurrence data collected systematically in several agricultural habitats to test how the predictive performance of SDMs for wild bee species depends on species traits, habitat type, and sampling technique. We used a species distribution modeling approach parametrized for the Netherlands, with presence records from 1990 to 2010 for 193 Dutch wild bees. For each species, we built a Maxent model based on 13 climate and landscape variables. We tested the predictive performance of the SDMs with independent datasets collected from orchards and arable fields across the Netherlands from 2010 to 2013, using transect surveys or pan traps. Model predictive performance depended on species traits and habitat type. Occurrence of bee species specialized in habitat and diet was better predicted than generalist bees. Predictions of habitat suitability were also more precise for habitats that are temporally more stable (orchards) than for habitats that suffer regular alterations (arable), particularly for small, solitary bees. As a conservation tool, SDMs are best suited to modeling rarer, specialist species than more generalist and will work best in long-term stable habitats. The variability of complex, short-term habitats is difficult to capture in such models and historical land use generally has low thematic resolution. To improve SDMs’ usefulness, models require explanatory variables and collection data that include detailed landscape characteristics, for example, variability of crops and flower availability. Additionally, testing SDMs with field surveys should involve multiple collection techniques.

Towards predictive understanding of regional climate change

Regional information on climate change is urgently needed but often deemed unreliable. To achieve credible regional climate projections, it is essential to understand underlying physical processes, reduce model biases and evaluate their impact on projections, and adequately account for internal variability. In the tropics, where atmospheric internal variability is small compared with the forced change, advancing our understanding of the coupling between long-term changes in upper-ocean temperature and the atmospheric circulation will help most to narrow the uncertainty. In the extratropics, relatively large internal variability introduces substantial uncertainty, while exacerbating risks associated with extreme events. Large ensemble simulations are essential to estimate the probabilistic distribution of climate change on regional scales. Regional models inherit atmospheric circulation uncertainty from global models and do not automatically solve the problem of regional climate change. We conclude that the current priority is to understand and reduce uncertainties on scales greater than 100 km to aid assessments at finer scales.