More Like this

1. Towards an Integrated Sustainability Evaluation of Energy Scenarios with Automated Information Exchange [Towards an Integrated Sustainability Evaluation of Energy Scenarios with Automated Information Exchange]

   https://doi.org/10.5220/0006302101880199  

   Schwarz, Jan Sören; Witt, Tobias; Nieße, Astrid; Geldermann, Jutta; Lehnhoff, Sebastian; Sonnenschein, Michael (January 2017, Proceedings of the 6th International Conference on Smart Cities and Green ICT Systems)

   To reshape energy systems towards renewable energy resources, decision makers need to decide today on how to make the transition. Energy scenarios are widely used to guide decision making in this context. While considerable effort has been put into developing energy scenarios, researchers have pointed out three requirements for energy scenarios that are not fulfilled satisfactorily yet: The development and evaluation of energy scenarios should (1) incorporate the concept of sustainability, (2) provide decision support in a transparent way and (3) be replicable for other researchers. To meet these requirements, we combine different methodological approaches: story-and-simulation (SAS) scenarios, multi-criteria decision-making (MCDM), information modeling and co-simulation. We show in this paper how the combination of these methods can lead to an integrated approach for sustainability evaluation of energy scenarios with automated information exchange. Our approach consists of a sustainability evaluation process (SEP) and an information model for modeling dependencies. The objectives are to guide decisions towards sustainable development of the energy sector and to make the scenario and decision support processes more transparent for both decision makers and researchers. 

   more » « less
2. Using Word Embeddings to Learn a Better Food Ontology

   https://doi.org/10.3389/frai.2020.584784  

   Youn, Jason; Naravane, Tarini; Tagkopoulos, Ilias (November 2020, Frontiers in Artificial Intelligence)

   null (Ed.)

   Food ontologies require significant effort to create and maintain as they involve manual and time-consuming tasks, often with limited alignment to the underlying food science knowledge. We propose a semi-supervised framework for the automated ontology population from an existing ontology scaffold by using word embeddings. Having applied this on the domain of food and subsequent evaluation against an expert-curated ontology, FoodOn, we observe that the food word embeddings capture the latent relationships and characteristics of foods. The resulting ontology, which utilizes word embeddings trained from the Wikipedia corpus, has an improvement of 89.7% in precision when compared to the expert-curated ontology FoodOn (0.34 vs. 0.18, respectively, p value = 2.6 × 10 –138 ), and it has a 43.6% shorter path distance (hops) between predicted and actual food instances (2.91 vs. 5.16, respectively, p value = 4.7 × 10 –84 ) when compared to other methods. This work demonstrates how high-dimensional representations of food can be used to populate ontologies and paves the way for learning ontologies that integrate contextual information from a variety of sources and types. 

   more » « less
3. A machine learning-based approach for building code requirement hierarchy extraction

   Zhang, Ruichuan; El-Gohary, Nora (January 2019, 2019 CSCE Annual Conference)

   Most of the existing automated code compliance checking (ACC) methods are unable to fully automatically convert complex building-code requirements into computer-processable forms. Such complex requirements usually have hierarchically complex clause and sentence structures. There is, thus, a need to decompose such complex requirements into hierarchies of much smaller, manageable requirement units that would be processable using most of the existing ACC methods. Rule-based methods have been used to deal with such complex requirements and have achieved high performance. However, they lack scalability, because the rules are developed manually and need to be updated and/or adapted when applied to a different type of building code. More research is, thus, needed to develop a scalable method to automatically convert the complex requirements into hierarchies of requirement units to facilitate the succeeding steps of ACC such as information extraction and compliance reasoning. To address this need, this paper proposes a new, machine learning-based method to automatically extract requirement hierarchies from building codes. The proposed method consists of five main steps: (1) data preparation and preprocessing; (2) data adaptation; (3) deep neural network model training for dependency parsing; (4) automated requirement segmentation and restriction interpretation based on the extracted dependencies; and (5) evaluation. The proposed method was trained using the English Treebank data; and was tested on sentences from the 2009 International Building Code (IBC) and the Champaign 2015 IBC Amendments. The preliminary results show that the proposed method achieved an average normalized edit distance of 0.32, a precision of 89%, a recall of 76%, and an F1-measure of 82%, which indicates good requirement hierarchy extraction performance. 

   more » « less
4. That’s the Wrong Lung! Evaluating and Improving the Interpretability of Unsupervised Multimodal Encoders for Medical Data

   McInerney, Jered; Young, Geoffrey; van de Meent, Jan-Willem; Wallace, Byron C. (January 2022, Proceedings of the Conference on Empirical Methods in Natural Language Processing)

   Pretraining multimodal models on Electronic Health Records (EHRs) provides a means of learning representations that can transfer to downstream tasks with minimal supervision. Recent multimodal models induce soft local alignments between image regions and sentences. This is of particular interest in the medical domain, where alignments might highlight regions in an image relevant to specific phenomena described in free-text. While past work has suggested that attention “heatmaps” can be interpreted in this manner, there has been little evaluation of such alignments. We compare alignments from a state-of-the-art multimodal (image and text) model for EHR with human annotations that link image regions to sentences. Our main finding is that the text has an often weak or unintuitive influence on attention; alignments do not consistently reflect basic anatomical information. Moreover, synthetic modifications — such as substituting “left” for “right” — do not substantially influence highlights. Simple techniques such as allowing the model to opt out of attending to the image and few-shot finetuning show promise in terms of their ability to improve alignments with very little or no supervision. We make our code and checkpoints open-source. 

   more » « less
5. Harmonizing Privacy Regarding Data Retention and Purging

   https://doi.org/10.1145/3538712.3538718  

   Scope, Nick; Rasin, Alexander; Lenard, Ben; Heart, Karen; Wagner, James (July 2022, Harmonizing Privacy Regarding Data Retention and Purging)

   Data privacy requirements are a complex and quickly evolving part of the data management domain. Especially in Healthcare (e.g., United States Health Insurance Portability and Accountability Act and Veterans Affairs requirements), there has been a strong emphasis on data privacy and protection. Data storage is governed by multiple sources of policy requirements, including internal policies and legal requirements imposed by external governing organizations. Within a database, a single value can be subject to multiple requirements on how long it must be preserved and when it must be irrecoverably destroyed. This often results in a complex set of overlapping and potentially conflicting policies. Existing storage systems are lacking sufficient support functionality for these critical and evolving rules, making compliance an underdeveloped aspect of data management. As a result, many organizations must implement manual ad-hoc solutions to ensure compliance. As long as organizations depend on manual approaches, there is an increased risk of non-compliance and threat to customer data privacy. In this paper, we detail and implement an automated comprehensive data management compliance framework facilitating retention and purging compliance within a database management system. This framework can be integrated into existing databases without requiring changes to existing business processes. Our proposed implementation uses SQL to set policies and automate compliance. We validate this framework on a Postgres database, and measure the factors that contribute to our reasonable performance overhead (13% in a simulated real-world workload). 

   more » « less