An official website of the United States government
Abstract Metaphor generation is both a creative act and a means of learning. When learning a new concept, people often create a metaphor to connect the new concept to existing knowledge. Does the manner in which people generate a metaphor, via sudden insight (Aha! moment) or deliberate analysis, influence the quality of generation and subsequent learning outcomes? According to some research, deliberate processing enhances knowledge retention; hence, generation via analysis likely leads to better concept learning. However, other research has shown that solutions generated via insight are better remembered. In the current study, participants were presented with science concepts and descriptions, then generated metaphors for the concepts. They also indicated how they generated each metaphor and rated their metaphor for novelty and aptness. We assessed participants’ learning outcomes with a memory test and evaluated the creative quality of the metaphors based on self‐ and crowd‐sourced ratings. Consistent with the deliberate processing benefit, participants became more familiar with the target science concept if they previously generated a metaphor for the concept via analysis compared to via insight. We also found that metaphors generated via analysis did not differ from metaphors generated via insight in quality (aptness or novelty) nor in how well they were remembered. However, participants’ self‐evaluations of metaphors generated via insight showed more agreement with independent raters, suggesting the role of insight in modulating the creative ideation process. These preliminary findings have implications for understanding the nature of insight during idea generation and its impact on learning.
more »
« less
Metaphor identification in cybersecurity texts: a lightweight linguistic approach
Abstract The use of metaphor in cybersecurity discourse has become a topic of interest because of its ability to aid communication about abstract security concepts. In this paper, we borrow from existing metaphor identification algorithms and general theories to create a lightweight metaphor identification algorithm, which uses only one external source of knowledge. The algorithm also introduces a real time corpus builder for extracting collocates; this is, identifying words that appear together more frequently than chance. We implement several variations of the introduced algorithm and empirically evaluate the output using the TroFi dataset, a de facto evaluation dataset in metaphor research. We find first, contrary to our expectation, that adding word sense disambiguation to our metaphor identification algorithm decreases its performance. Second, we find, that our lightweight algorithms perform comparably to their existing, more complex, counterparts. Finally, we present the results of several case studies to observe the utility of the algorithm for future research in linguistic metaphor identification in text related to cybersecurity texts and threats.
more »
« less
- Award ID(s):
- 1564293
- PAR ID:
- 10362090
- Publisher / Repository:
- Springer Science + Business Media
- Date Published:
- Journal Name:
- SN Applied Sciences
- Volume:
- 4
- Issue:
- 2
- ISSN:
- 2523-3963
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
The introduction of deep learning and CNNs to image recognition problems has led to state-of-the-art classification accuracy. However, CNNs exacerbate the issue of algorithm explainability due to deep learning’s black box nature. Numerous explainable AI (XAI) algorithms have been developed that provide developers insight into the operations of deep learning. We aim to make XAI explanations more user-centric by introducing modifications to existing XAI algorithms based on cognitive theory. The goal of this research is to yield intuitive XAI explanations that more closely resemble explanations given by experts in the domain of bird watching. Using an existing base XAI algorithm, we conducted two user studies with expert bird watchers and found that our novel averaged and contrasting XAI algorithms are significantly preferred over the base XAI algorithm for bird identification.more » « less
-
10.17605/OSF.IO/AT4XE Despite increased use of digital biodiversity data in research, reliable methods to identify datasets are not widely adopted. While commonly used location-based dataset identifiers such as URLs help to easily download data today, additional identification schemes are needed to ensure long term access to datasets. We propose to augment existing location- and DOI-based identification schemes with cryptographic content-based identifiers. These content-based identifiers can be calculated from the datasets themselves using available cryptographic hashing algorithms (e.g., sha256). These algorithms take only the digital content as input to generate a unique identifier without needing a centralized identification administration. The use of content-based identifiers is not new, but a re-application of change management techniques used in the popular version control system "git". We show how content-based identifiers can be used to version datasets, to track the dataset locations, to monitor their reliability, and to efficiently detect dataset changes. We discuss the results of using our approach on datasets registered in GBIF and iDigBio from Sept 2018 to May 2020. Also, we propose how reliable, decentralized, dataset indexing and archiving systems can be devised. Lastly, we outline a modification to existing data citation practices to help work towards more reproducible and reusable research workflows.more » « less
-
Abstract In this paper, we address several facets of the problem we call local government cyber insecurity—a problem that plagues such governments across the nation, if not the world. We describe this problem and discuss its manifestations in local governments. This is followed by our analysis of why, on average, local government cybersecurity is managed and practiced so poorly. Next, we discuss several constraints on local governments that may help to explain why so many of these governments are not able to provide highly effective cybersecurity. We then discuss steps that local governments can and should take to improve their cybersecurity, including adopting dedicated cybersecurity budgets, adopting several highly recommended cybersecurity policies, and following best cybersecurity practices.more » « less
-
We propose a new variant of the top arm identification problem, top feasible arm identification, where there are K arms associated with D-dimensional distributions and the goal is to find m arms that maximize some known linear function of their means subject to the constraint that their means belong to a given set P € R D. This problem has many applications since in many settings, feedback is multi-dimensional and it is of interest to perform constrained maximization. We present problem-dependent lower bounds for top feasible arm identification and upper bounds for several algorithms. Our most broadly applicable algorithm, TF-LUCB-B, has an upper bound that is loose by a factor of OpDlogpKqq. Many problems of practical interest are two dimensional and, for these, it is loose by a factor of OplogpKqq. Finally, we conduct experiments on synthetic and real-world datasets that demonstrate the effectiveness of our algorithms. Our algorithms are superior both in theory and in practice to a naive two-stage algorithm that first identifies the feasible arms and then applies a best arm identification algorithm to the feasible arms.more » « less
