Learning symbolic formulations in design: Syntax, semantics, and knowledge reification

2010 ◽  
Vol 24 (1) ◽  
pp. 63-85 ◽  
Author(s):  
Somwrita Sarkar ◽  
Andy Dong ◽  
John S. Gero
Keyword(s):  
Language Processing ◽  
Design Problem ◽  
Knowledge Engineering ◽  
Hydraulic Cylinder ◽  
Semantic Knowledge ◽  
Design Structure ◽  
Statistical Natural Language Processing ◽  
Design Representations ◽  
And Performance ◽  
Value Decomposition

AbstractAn artificial intelligence (AI) algorithm to automate symbolic design reformulation is an enduring challenge in design automation. Existing research shows that design tools either require high levels of knowledge engineering or large databases of training cases. To address these limitations, we present a singular value decomposition (SVD) and unsupervised clustering-based method that performs design reformulation by acquiring semantic knowledge from the syntax of design representations. The development of the method was analogically inspired by applications of SVD in statistical natural language processing and digital image processing. We demonstrate our method on an analytically formulated hydraulic cylinder design problem and an aeroengine design problem formulated using a nonanalytic design structure matrix form. Our results show that the method automates various design reformulation tasks on problems of varying sizes from different design domains, stated in analytic and nonanalytic representational forms. The behavior of the method presents observations that cannot be explained by pure symbolic AI approaches, including uncovering patterns of implicit knowledge that are not readily encoded as logical rules, and automating tasks that require the associative transformation of sets of inputs to experiences. As an explanation, we relate the structure and performance of our algorithm with findings in cognitive neuroscience, and present a set of theoretical postulates addressing an alternate perspective on how symbols may interact with each other in experiences to reify semantic knowledge in design representations.

scholarly journals Emotion Recognition on Edge Devices: Training and Deployment

Sensors ◽  
2021 ◽  
Vol 21 (13) ◽  
pp. 4496
Author(s):  
Vlad Pandelea ◽  
Edoardo Ragusa ◽  
Tommaso Apicella ◽  
Paolo Gastaldo ◽  
Erik Cambria
Keyword(s):  
Emotion Recognition ◽  
Language Processing ◽  
Computational Cost ◽  
Sequential Learning ◽  
High Quality ◽  
Fast Training ◽  
Online Sequential Learning ◽  
And Performance ◽  
Resource Constrained Devices ◽  
Constrained Devices

Emotion recognition, among other natural language processing tasks, has greatly benefited from the use of large transformer models. Deploying these models on resource-constrained devices, however, is a major challenge due to their computational cost. In this paper, we show that the combination of large transformers, as high-quality feature extractors, and simple hardware-friendly classifiers based on linear separators can achieve competitive performance while allowing real-time inference and fast training. Various solutions including batch and Online Sequential Learning are analyzed. Additionally, our experiments show that latency and performance can be further improved via dimensionality reduction and pre-training, respectively. The resulting system is implemented on two types of edge device, namely an edge accelerator and two smartphones.

scholarly journals Google Play Content Scraping and Knowledge Engineering using Natural Language Processing Techniques with the Analysis of User Reviews

2020 ◽  
Vol 30 (1) ◽  
pp. 192-208 ◽  
Author(s):  
Hamza Aldabbas ◽  
Abdullah Bajahzar ◽  
Meshrif Alruily ◽  
Ali Adil Qureshi ◽  
Rana M. Amir Latif ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Language Processing ◽  
Mobile Application ◽  
Knowledge Engineering ◽  
Machine Learning Algorithms ◽  
Application Development ◽  
User Reviews ◽  
N Gram ◽  
Logistic Regression Algorithm ◽  
Google Play

Abstract To maintain the competitive edge and evaluating the needs of the quality app is in the mobile application market. The user’s feedback on these applications plays an essential role in the mobile application development industry. The rapid growth of web technology gave people an opportunity to interact and express their review, rate and share their feedback about applications. In this paper we have scrapped 506259 of user reviews and applications rate from Google Play Store from 14 different categories. The statistical information was measured in the results using different of common machine learning algorithms such as the Logistic Regression, Random Forest Classifier, and Multinomial Naïve Bayes. Different parameters including the accuracy, precision, recall, and F1 score were used to evaluate Bigram, Trigram, and N-gram, and the statistical result of these algorithms was compared. The analysis of each algorithm, one by one, is performed, and the result has been evaluated. It is concluded that logistic regression is the best algorithm for review analysis of the Google Play Store applications. The results have been checked scientifically, and it is found that the accuracy of the logistic regression algorithm for analyzing different reviews based on three classes, i.e., positive, negative, and neutral.

Performance Oriented Design of Semiregular Anisotropic Porous Structures

2018 ◽  
Author(s):  
Chao Xu ◽  
Lili Pan ◽  
Ming Li ◽  
Shuming Gao
Keyword(s):  
3D Printing ◽  
Porous Materials ◽  
Design Problem ◽  
Degrees Of Freedom ◽  
Design Space ◽  
Manufacturing Technology ◽  
Porous Structures ◽  
Precise Control ◽  
And Performance

Porous materials / structures have wide applications in industry, since the sizes, shapes and positions of their pores can be adjusted on various demands. However, the precise control and performance oriented design of porous structures are still urgent and challenging, especially when the manufacturing technology is well developed due to 3D printing. In this study, the control and design of anisotropic porous structures are studied with more degrees of freedom than isotropic structures, and can achieve more complex mechanical goals. The proposed approach introduces Super Formula to represent the structural cells, maps the design problem to an optimal problem using PGD, and solves the optimal problem using MMA to obtain the structure with desired performance. The proposed approach is also tested on the performance of the expansion of design space, the capture of the physical orientation and so on.

scholarly journals Control and Performance Analysis of a Digital Direct Hydraulic Cylinder Drive

2020 ◽  
Author(s):  
Niels H Pedersen ◽  
Per Johansen ◽  
Lasse Schmidt ◽  
Rudolf Scheidl ◽  
Torben O. Andersen
Keyword(s):  
Motor Unit ◽  
System Analysis ◽  
Hydraulic Cylinder ◽  
Controller Synthesis ◽  
Successful Implementation ◽  
Discrete Elements ◽  
Hydraulic Pump ◽  
And Performance ◽  
Feedback Control Strategy ◽  
Novel Concept

This paper concerns control of a digital direct hydraulic cylinder drive (D-DHCD) and is a novel concept with the potential to become the future solution for energy efficient hydraulic drives. The concept relies on direct control of a differential cylinder by a single hydraulic pump/motor unit connected to each cylinder inlet/outlet. The pump/motor unit in this research uses the digital displacement technology and comprises of numerous individually digital controlled pressure chambers, such that the ratio of active (motoring, pumping or idling) chambers determines the machine power throughput. This feature reduces energy losses to a minimum, since the inactive (idling) chambers has very low losses. A single DDM may provide individually load control for several cylinders without excessive throttling due to various load sizes. Successful implementation of the concept relies on proper control of the DDM, which demands a dynamical model that allows for system analysis and controller synthesis. This is a challenging task, due to the highly non-smooth machine behavior, comprising both non-linear continuous and discrete elements. This paper presents the first feedback control strategy for a D-DHCD concept, based on a discrete dynamical approximation and investigates the control performance in a mathematical simulation model representing the physical system.

A-2 Clustering and Switching Indices as Discriminators between Alzheimer’s Dementia, Vascular Dementia, and Mild Cognitive Impairment on a Semantic Fluency Task

2021 ◽  
Vol 36 (6) ◽  
pp. 1023-1023
Author(s):  
Amanda M Wisinger ◽  
Matthew S Phillips ◽  
Dustin A Carter ◽  
Kyle J Jennette ◽  
Joseph W Fink
Keyword(s):  
Cognitive Impairment ◽  
Mild Cognitive Impairment ◽  
Executive Functioning ◽  
Vascular Dementia ◽  
Cluster Size ◽  
Semantic Knowledge ◽  
Group Differences ◽  
Semantic Fluency ◽  
And Performance ◽  
Memory Concerns

Abstract Objective Studies that have used semantic fluency tasks to guide differential diagnosis of Alzheimer’s disease (ad) and vascular dementia (VaD) typically only examine the total number of words produced, which has yielded conflicting results. The present study examined whether other indices of semantic fluency (i.e., clustering and switching), which are thought to better isolate the components of semantic memory and executive functioning abilities, would discriminate among ad, VaD, and mild cognitive impairment (MCI). Method A retrospective sample of 156 patients (mean age = 78.64; 76.3% female, 23.7% male; 26.9% White, 71.2% Black, 1.9% Other) who completed a comprehensive neuropsychological evaluation as part of a workup related to memory concerns and were diagnosed with ad, VaD, or MCI was utilized. Separate univariate analyses of variance were used to examine group differences on three indices of semantic fluency (animals): total words, mean cluster size, and number of switches. Results There was a significant main effect of group for total words [F(2,153) = 7.09, p = 0.001], mean cluster size [F(2, 153) = 3.44, p = 0.035] and number of switches [F(2,153) = 3.36, p = 0.037]. Bonferroni post-hoc tests revealed that the ad and VaD groups produced significantly fewer total words than the MCI group, the ad group produced significantly smaller clusters than the VaD group, and the VaD group produced significantly fewer switches than the MCI group. Conclusion Observed group differences suggest that clustering and switching may aid in discriminating between dementia etiologies. Future studies may benefit from examining the association between these fluency indices and performance on executive functioning and semantic knowledge tasks to better understand these findings.

Sign in / Sign up

Export Citation Format

Share Document