35180 research outputs found
Sort by
King South County Resource Guide - Washington
Get PDFCounty level and region-specific resource guides focused on mental health and substance use
Mason County Resource Guide - Washington
Get PDFCounty level and region-specific resource guides focused on mental health and substance use
Franklin County Resource Guide - Idaho
Get PDFCounty level and region-specific resource guides focused on mental health and substance use
Boise County Resource Guide - Idaho
Get PDFCounty level and region-specific resource guides focused on mental health and substance use
Idaho County Resource Guide - Idaho
Get PDFCounty level and region-specific resource guides focused on mental health and substance use
Skagit County Resource Guide - Washington
Get PDFCounty level and region-specific resource guides focused on mental health and substance use
Exploring the Future of Library Cataloging with AI and Multilingual Embeddings
Get PDFFacing the dual challenges of library staffing shortages and the complexity of cataloging Chinese materials, we set out to create a practical solution. In this post, we introduce a multilingual subject analysis tool we’ve developed — combining large language models, vector databases, and multilingual embeddings
Habitat Mosaics in Dynamic Landscapes Woodland Caribou in the Northwestern United States
Get PDFBecause of rapid human encroachment and climate change, northern boreal forests (also referred to as Taiga) are one of the biomes most rapidly experiencing changes in ecological composition and biodiversity. One of the changes impacting boreal-adapted species most substantially is the gradual shift of suitable habitat. This shift has been seen across taxa and has been suggested to have been a contributing factor to the extirpation of groups such as the South Selkirk Mountain population of woodland caribou (Rangifer tarandus caribou). Quantifying the changes in this species’ suitable habitat and determining the suitability of remaining habitat in the species’ former range, may provide insight into what conservation actions would be most beneficial to mitigating the effects of habitat degradation. Based upon recent assessments of caribou habitat in British Columbia I hypothesized that semi-suitable habitat patches (defined as ≥50% on the Habitat Suitability Index created in this study) would be more abundant and larger in area than highly suitable patches (≥75% on the Habitat Suitability Index), that highly suitable patches would be limited by size (1800m), and that the semi-suitable habitat would have greater impact on overall landscape connectivity metrics than highly-suitable patches due to their larger size and increased abundance.After conducting a thorough literature review, I acquired spatially explicit data (i.e., Geographic Information system layers) for 12 habitat components that have been identified as promoting habitat use by woodland caribou: mean annual temperature, precipitation, and snow depth; slope; forest age; distance from water sources; distance from road, rails and frequented trails; distance from urban areas; and distance from agriculture. After standardizing each dataset, I developed a habitat suitability function that provided relative weights for each of these components and displayed them across four national forests (NF: Colville, Coeur d’Alene, Kaniksu, and Kootenai) to create a map of habitat suitability (referred to as the habitat suitability index or HSI). This analysis indicated that suitable caribou habitat was found in many small patches dispersed throughout the landscape. Based on this result, I chose to examine patch connectivity to make more effective proposals for future management.Using the HSI, I separated habitat patches into either semi-suitable or highly suitable categories and gathered information for both categories on metrics such as average patch size and total amount of habitat area. I then identified ten patches as potential focal patches for woodland caribou conservation: the five largest semi-suitable patches and the five largest highly suitable patches. I then created buffers of 3km, 5km and 15km around each patch to assess each patch’s ability to increase total habitat connectivity and increase in total habitat area based upon the likelihood of caribou dispersal at these distances. I used a in a One-Way ANOVA to examine the relationship between suitability category and connectivity metric. Finally, after quantifying the connectivity metric values, the ten patches were then ranked for overall influence on habitat connectivity.Most focal patches were identified in regions south of the last known location of woodland caribou from the Selkirk population. As I hypothesized, semi-suitable patches were larger and more abundant compared to highly suitable patches. Semi-suitable habitat patches had a significantly greater impact on the increase in total patch area, as well as the number of incorporated patches across all buffer distances. By increasing the connectivity and land protection of this region, potentially through a protected area network, future conservation efforts for woodland caribou may see greater success
CONFORMALIZED UNCERTAINTY REGIONS FOR MACHINE LEARNING-BASED MULTIPLE COGNITIVE HEALTH MEASURES FROM SMART WATCH SENSOR DATA
Get PDFMachine learning (ML) algorithms play an increasingly critical role in remote health monitoring, automating health assessments, and extending medical professionals’ reach incaring for an aging population. However, ML models often provide predictions without quantifying uncertainty, which is essential for safe deployment in healthcare. This paperpresents a novel method referred to as Uncertainty Regions via Importance-weighted Calibration (URIC) for automating the uncertainty-based prediction of multiple clinical cognitive health measures from continuous smartwatch sensor data. URIC leverages Conformal Prediction (CP), a rigorous uncertainty quantification (UQ) framework that guarantees user-defined coverage (e.g., ground truth is in the predicted region with 95% probability). The key innovation of URIC is constructing smaller, more interpretable prediction regions by assigning importance weights to calibration examples. It forms prediction regions by using the most relevant calibration examples, ensuring tight regions without sacrificing coverage. We evaluate URIC for predicting the cognitive health measures of 157 adults who were either cognitively healthy or experienced mild cognitive impairment (MCI). Compared to other UQ methods, our method constructs smaller prediction regions across all multi-target combinations while maintaining the user-specified coverage. This approach can be integrated into human-ML collaborative systems to improve diagnostic accuracy and interpretability in sensitive multi-target healthcare tasks
Generative Artificial Intelligence in Pre-Service Teacher Education: A Systematic Review of Applications, Outcomes, and Implications
Get PDFThis study is guided by the following research questions.1. Which Generative AI has been used to support pre-service teachers in teacher education?2. What research designs and measures are employed?3. What outcomes (cognitive, affective, professional) are reported?4. What challenges (ethical issues) are reported