An introduction to neural networks and their applications in manufacturing

Wu, Bin

doi:10.1007/bf01473534

Cited by 28 publications

(7 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We observe that words with similar semantic meaning are closer in the word embedding space, e.g., {run, walk, move}, whereas words having different semantic meaning are far, e.g., run and eat. In Natural Language Processing (NLP), researchers use techniques such as neural network [64] and co-occurrence matrix [43] to project words onto a k-dimensional vector space, and words appearing in similar textual contexts get closer embedding in the vector space. One of the most popular method to extract word embedding from a large scale corpus is Word2Vec [38], which uses a two layer neural network to predict the surrounding words of a target word.…”

Section: Word Embeddingmentioning

confidence: 99%

Wi-Fringe: Leveraging Text Semantics in WiFi CSI-Based Device-Free Named Gesture Recognition

Islam¹,

Nirjon²

2019

Preprint

View full text Add to dashboard Cite

The lack of adequate training data is one of the major hurdles in WiFi-based activity recognition systems. In this paper, we propose Wi-Fringe, which is a WiFi CSI-based device-free human gesture recognition system that recognizes named gestures, i.e., activities and gestures that have a semantically meaningful name in English language, as opposed to arbitrary free-form gestures. Given a list of activities (only their names in English text), along with zero or more training examples (WiFi CSI values) per activity, Wi-Fringeis able to detect all activities at runtime. In other words, a subset of activities that Wi-Fringedetects do not require any training examples at all. This is achieved by leveraging prior knowledge of these activities from another domain, i.e., text. We show for the first time that by utilizing the state-of-the-art semantic representation of English words, which is learned from a massive dataset such as the Wikipedia (e.g., Google's word-to-vector [38]) and verb attributes learned from how a word is defined (e.g, American Heritage Dictionary), we can enhance the capability of WiFi-based named gesture recognition systems that lack adequate training examples per class. We propose a novel cross-domain knowledge transfer algorithm between radio frequency (RF) and text to lessen the burden on developers and end-users from the tedious task of data collection for all possible activities. To evaluate Wi-Fringe, we collect data from four volunteers for 20 activities in two environments. We show that Wi-Fringeachieves an accuracy of up to 90% for two unseen activities and 61% for up to six unseen activities.

show abstract

Section: Word Embeddingmentioning

confidence: 99%

Wi-Fringe: Leveraging Text Semantics in WiFi CSI-Based Device-Free Named Gesture Recognition

Islam¹,

Nirjon²

2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Per their report, the neural networks were capable of detecting tool condition accurately. Wu (1992) developed a neural network model to detect tool failure based on the level of cutting force and vibration or acoustic emission. Brophy et al (2002) proposed a two-stage neural network model to detect anomalies in the drilling process.…”

Section: Condition Monitoringmentioning

confidence: 99%

Artificial Neural Networks

Sarker

2006

Artificial Neural Networks in Finance and Manufacturing

View full text Add to dashboard Cite

The primary aim of this chapter is to present an overview of the artificial neural network basics and operation, architectures, and the major algorithms used for training the neural network models. As can be seen in subsequent chapters, neural networks have made many useful contributions to solve theoretical and practical problems in finance and manufacturing areas. The secondary aim here is therefore to provide a brief review of artificial neural network applications in finance and manufacturing areas.

show abstract

“…ANNs have been used to solve numerous problems associated with manufacturing operations. Application of ANN in manufacturing system design [2]- [6], manufacturing process control [7], robot scheduling [8], industrial pattern recognition [9], and manufacturing operational decision [10] is reported.…”

Section: Introductionmentioning

confidence: 99%

Application of a Coupled Simulation-Optimization System Called AnyPLOS in a Cold Foam Production Line

Jahanpour¹,

Tahouri²

2013

IJMO

View full text Add to dashboard Cite

Abstract-A prototype simulation-optimization system called AnyPLOS, which couples an Artificial Neural Network (ANN) based simulation model with a genetic algorithm optimization model, is presented. AnyPLOS is designed to discover value of effective input parameters of a production line so that all required quality control tests on the output product is satisfied. First an ANN was trained and tested to provide an acceptable level of accuracy in prediction of production line outputs, and then it was coupled with a GA optimization module to find desired solutions. A real world case study, in Erish Khodro manufacturing company, was set up and the foam production process input parameters were optimized so that the produced samples satisfied quality requirements. In order to verify the results, discovered solutions were used to produce real foam samples in the production line. After that, quality control tests were performed on samples. Quality test results were, as predicted by ANN, within the desired range. In order to estimate the performance of the trained ANN, experimental observations were compared to values which were predicted by ANN. A convincing correlation was found between ANN predictions and experimental values.

show abstract

An introduction to neural networks and their applications in manufacturing

Cited by 28 publications

References 4 publications

Wi-Fringe: Leveraging Text Semantics in WiFi CSI-Based Device-Free Named Gesture Recognition

Wi-Fringe: Leveraging Text Semantics in WiFi CSI-Based Device-Free Named Gesture Recognition

Artificial Neural Networks

Application of a Coupled Simulation-Optimization System Called AnyPLOS in a Cold Foam Production Line

Contact Info

Product

Resources

About