Building Features in Schools That Influence Academic Performance

School decision makers are faced with a great many decisions when considering a school renovation or new school building. All stakeholders want a building that is safe and provides an optimal learning environment. However, it is often difficult to know which building features will have the greatest effect on student learning. Because of a limited understanding of the relationship between individual building features and student learning, researchers at the University of Oklahoma hope to explore how building components influence student and teacher performance. This paper explores the importance of school building features that can be designed and changed during a renovation project. The hope is to one day determine which features have the greatest impact on student test scores. The research team believes that although it is difficult to find the exact relationship between each building features and student outcomes with one study, if multiple users repeat the same or similar studies, hopefully we will one day know the effect of these building features. In order to develop feature building users survey and physical assessment tools, it was necessary for investigators to develop a list of important building features and their associated definitions in layman terms. This was accomplished through utilization and conducting of a CAB （community advisory board） and subject matter expert materials. In addition, previous research relating to different school building features and their associations with student performance were reviewed. To define and narrow the list the researchers, community educational, and building professionals rated based on their professional experience, how directly related each feature is to student performance. The building feature list serves as a starting point to determine which features should be analyzed in a later phase of the project. It is hoped that resulting tools based on the work of this project can be used by school decision-makers and researchers to access building features that have been identified through research as being important for student and teacher performance.

Unsupervised learning of load signatures to estimate energy-related building features using surrogate modelling techniques

Characterization of an existing building’s energy-related features is critical to inform maintenance and retrofit decisions.However,existing field-scale characterization methods tend to be labour intensive,invasive,and require high fidelity longitudinal data gathered through tightly regulated experiments.This highlights the need for a low cost,scalable,and efficient screening method.This paper puts forward a surrogate model-based approach to rapidly estimate energy-related building features.To this end,EnergyPlus models for 12 midrise office archetypes,all with a rectangular footprint,are developed.Ten thousand variants of each archetype are generated by altering envelope,causal heat gain,and heating,ventilation,and air conditioning operation features.A unique load signature is derived for each variant’s heating and cooling energy use.The parameters of the load signatures are clustered,then each cluster is associated with a set of plausible energy-related features.The accuracy of the results was evaluated using five test buildings not seen by the algorithm.The method could effectively identify building features with reasonable accuracy and no significant degradation in performance across all 12 archetypes.

Detecting Malicious Uniform Resource Locators Using an Applied Intelligence Framework

The potential of text analytics is revealed by Machine Learning(ML)and Natural Language Processing(NLP)techniques.In this paper,we propose an NLP framework that is applied to multiple datasets to detect malicious Uniform Resource Locators(URLs).Three categories of features,both ML and Deep Learning(DL)algorithms and a ranking schema are included in the proposed framework.We apply frequency and prediction-based embeddings,such as hash vectorizer,Term Frequency-Inverse Dense Frequency(TF-IDF)and predictors,word to vector-word2vec(continuous bag of words,skip-gram)from Google,to extract features from text.Further,we apply more state-of-the-art methods to create vectorized features,such as GloVe.Additionally,feature engineering that is specific to URL structure is deployed to detect scams and other threats.For framework assessment,four ranking indicators are weighted:computational time and performance as accuracy,F1 score and type error II.For the computational time,we propose a new metric-Feature Building Time(FBT)as the cutting-edge feature builders(like doc2vec or GloVe)require more time.By applying the proposed assessment step,the skip-gram algorithm of word2vec surpasses other feature builders in performance.Additionally,eXtreme Gradient Boost(XGB)outperforms other classifiers.With this setup,we attain an accuracy of 99.5%and an F1 score of 0.99.

Multiscale and multiphysics explorations of the transient deposition processes and additive characteristics during laser 3D printing

Laser directed energy deposition(DED)involves complex physical processes,and the trial and error examinations are time consuming and cost expensive.The research paradigm can be reshaped using advanced phenomenological models via computing the spatiotemporal variations of the build features.In this work,multi-layer and multi-track laser DED of Ti-6 Al-4 V were systematically explored on multiple scales including the 1D track,the 2D layer and the 3D full build considering the complex transport of energy,mass,and momentum in the moving freeform molten pool.The results showed that convex,nearflat,and wavy builds were generated using gradually larger hatch spacings.The profiles of individual tracks and layers were extracted through the unique advantages of the model.The individual tracks exhibited various patterns and rotated with specific inclinations to form distinct layer profiles.The net increments of the deposit generated upon the printing of a new track during the continuous deposition process showed that the smaller hatch spacing caused higher overlap rate of horizontally adjacent tracks but lower remelting rate of vertically adjacent tracks in neighboring layers.The 3D numerical model was validated with corresponding experiments for various process conditions.The scientific findings can provide useful insights for further researches of DED.