scholarly journals Evaluation of Goaf Stability Based on Transfer Learning Theory of Artificial Intelligence

IEEE Access ◽  
2019 ◽  
Vol 7 ◽  
pp. 96912-96925
Author(s):  
Yaguang Qin ◽  
Zhouquan Luo ◽  
Jie Wang ◽  
Shaowei Ma ◽  
Chundi Feng
Keyword(s):  
Artificial Intelligence ◽  
Transfer Learning ◽  
Learning Theory

scholarly journals Cloud based mobile solution for early detection of Skin Cancer using Artificial Intelligence

2021 ◽  
pp. 312-324
Author(s):  
Pawan Sonawane ◽  
Sahel Shardhul ◽  
Raju Mendhe
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Skin Cancer ◽  
Early Detection ◽  
Transfer Learning ◽  
Network Model ◽  
Neural Network Model ◽  
Mobile Application ◽  
Low Cost ◽  
Skin Tumors

The vast majority of skin cancer deaths are from melanoma, with about 1.04 million cases annually. Early detection of the same can be immensely helpful in order to try to cure it. But most of the diagnosis procedures are either extremely expensive or not available to a vast majority, as these centers are concentrated in urban regions only. Thus, there is a need for an application that can perform a quick, efficient, and low-cost diagnosis. Our solution proposes to build a server less mobile application on the AWS cloud that takes the images of potential skin tumors and classifies it as either Malignant or Benign. The classification would be carried out using a trained Convolution Neural Network model and Transfer learning (Inception v3). Several experiments will be performed based on Morphology and Color of the tumor to identify ideal parameters.

Artificial intelligence in the workplace – A double-edged sword

2020 ◽  
Vol 37 (5) ◽  
pp. 253-265
Author(s):  
Uta Wilkens
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Organizational Learning ◽  
Learning Theory ◽  
Practical Implication ◽  
Job Design ◽  
Macro Level ◽  
Learning Approaches ◽  
Content Type ◽  
Meso Level

PurposeThe aim of this paper is to outline how artificial intelligence (AI) can augment learning process in the workplace and where there are limitations.Design/methodology/approachThe paper is a theoretical-based outline with reference to individual and organizational learning theory, which are related to machine learning methods as they are currently in use in the workplace. Based on these theoretical insights, the paper presents a qualitative evaluation of the augmentation potential of AI to assist individual and organizational learning in the workplace.FindingsThe core outcome is that there is an augmentation potential of AI to enhance individual learning and development in the workplace, which however should not be overestimated. AI has a complementarity to individual intelligence, which can lead to an advancement, especially in quality, accuracy and precision. Moreover, AI has a potential to support individual competence development and organizational learning processes. However, a further outcome is that AI in the workplace is a double-edged sword, as it easily shows reinforcement effects in individual and organizational learning, which have a backside of unintended effects.Research limitations/implicationsThe conceptual outline makes use of examples for illustrating phenomenon but needs further empirical analysis. The research focus on the meso level of the workplace does not fully refer to macro level outcomes.Practical implicationsThe practical implication is that it is a matter of socio-technical job design to integrate AI in the workplace in a valuable manner. There is a need to keep the human-in-the-loop and to complement AI-based learning approaches with non-AI counterparts to reach augmentation.Originality/valueThe paper faces workplace learning from an interdisciplinary perspective and bridges insights from learning theory with methods from the machine learning community. It directs the social science discourse on AI, which is often on macro level to the meso level of the workplace and related issues for job design and therefore provides a complementary perspective.

scholarly journals RadImageNet: A Large-scale Radiologic Dataset for Enhancing Deep Learning Transfer Learning Research

2021 ◽  
Author(s):  
Yang Yang ◽  
Xueyan Mei ◽  
Philip Robson ◽  
Brett Marinelli ◽  
Mingqian Huang ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Medical Imaging ◽  
Transfer Learning ◽  
Large Scale ◽  
Medical Images ◽  
Bone Age ◽  
Natural World ◽  
X Rays ◽  
Starting Point ◽  
Size Diversity

Abstract Most current medical imaging Artificial Intelligence (AI) relies upon transfer learning using convolutional neural networks (CNNs) created using ImageNet, a large database of natural world images, including cats, dogs, and vehicles. Size, diversity, and similarity of the source data determine the success of the transfer learning on the target data. ImageNet is large and diverse, but there is a significant dissimilarity between its natural world images and medical images, leading Cheplygina to pose the question, “Why do we still use images of cats to help Artificial Intelligence interpret CAT scans?”. We present an equally large and diversified database, RadImageNet, consisting of 5 million annotated medical images consisting of CT, MRI, and ultrasound of musculoskeletal, neurologic, oncologic, gastrointestinal, endocrine, and pulmonary pathologies over 450,000 patients. The database is unprecedented in scale and breadth in the medical imaging field, constituting a more appropriate basis for medical imaging transfer learning applications. We found that RadImageNet transfer learning outperformed ImageNet in multiple independent applications, including improvements for bone age prediction from hand and wrist x-rays by 1.75 months (p<0.0001), pneumonia detection in ICU chest x-rays by 0.85% (p<0.0001), ACL tear detection on MRI by 10.72% (p<0.0001), SARS-CoV-2 detection on chest CT by 0.25% (p<0.0001) and hemorrhage detection on head CT by 0.13% (p<0.0001). The results indicate that our pre-trained models that are open-sourced on public domains will be a better starting point for transfer learning in radiologic imaging AI applications, including applications involving medical imaging modalities or anatomies not included in the RadImageNet database.

scholarly journals The Influences of CSR’s Multi-Dimensional Characteristics on Firm Value Determination by a Fusion Approach

2018 ◽  
Vol 10 (11) ◽  
pp. 3872 ◽  
Author(s):  
Kuang-Hua Hu ◽  
Sin-Jin Lin ◽  
Jau-Yang Liu ◽  
Fu-Hsiang Chen ◽  
Shih-Han Chen
Keyword(s):  
Artificial Intelligence ◽  
Sustainable Development ◽  
Corporate Social Responsibility ◽  
Learning Theory ◽  
Firm Value ◽  
Fusion Model ◽  
Value Determination ◽  
Corporate Social ◽  
The Relationship ◽  
Fusion Approach

Corporate social responsibility (CSR) implementation has been widely acknowledged as playing a key part in enhancing firm value as well as achieving sustainable development. However, up to now the extant works in the literature have yielded non-conclusive results regarding the relationships between CSR and firm value. One of the possible reasons is that the studies ignore the multi-dimensional characteristics of CSR—that is, they merely utilize a singular synthesized indicator as a proxy to represent the corporate’s CSR performance as being unreliable and problematic. Thus, this study breaks down CSR into numerous dimensions and further examines each dimension’s impact on firm value. By doing so, managers can allocate their firm’s valuable resources to suitable areas so as to increase its reputation and value. In addition, this research sets up an artificial intelligence (AI)-based fusion model, grounded by fusion learning theory that aims at complementing the error made by a singular model, to examine the relationship between CSR’s multidimensional characteristics and firm value. Through different combinations of adopted strategies, users can realize the most representative features from an over-abundant database.

Sign in / Sign up

Export Citation Format

Share Document