Software development is driven by software comprehension. Controlling a software development process is dependent on controlling software comprehension. Measures of factors that influence software comprehension are re...
详细信息
Software development is driven by software comprehension. Controlling a software development process is dependent on controlling software comprehension. Measures of factors that influence software comprehension are required in order to achieve control. The use of high-level languages results in many different kinds of lines of code that require different levels of comprehension effort. As the reader learns the set of arrangements of operators, attributes and labels particular to an application, comprehension is eased as familiar arrangements are repeated. Elements of cognition that describe the mechanics of comprehension serve as a guide to assessing comprehension demands in the understanding of programs written in high level languages. A new metric, kinds of lines of code identifier density is introduced and a case study demonstrates its application and importance. Related work is discussed.
In this paper, we formally present a methodology that, by using genetics, allows us to adapt the optimal intelligence of the elements conforming a living system. We call these environments meta-adaptable systems. We p...
详细信息
In this paper, we formally present a methodology that, by using genetics, allows us to adapt the optimal intelligence of the elements conforming a living system. We call these environments meta-adaptable systems. We propose systems in which the capability of the agents to adapt themselves to the environment (via intelligence) is itself adapted according to the changes of their own environment. We show a simple but illustrative example of such a meta-adaptable system, which can be easily extended to deal with more real situations as the construction of software systems based on intelligent mobile agents. This system has been fully implemented, and some interesting conclusions have been extracted from it.
Designing complex machines and systems that operate in very difficult remote locations, with largely unknown conditions, is very challenging. Specifications for such systems must be extremely detailed and extensive, w...
详细信息
Designing complex machines and systems that operate in very difficult remote locations, with largely unknown conditions, is very challenging. Specifications for such systems must be extremely detailed and extensive, with input from professionals who have designed such systems before, and who gained considerable experience from their operations. Since much of the operating environment is not not known in advance, cognitive informatics and computing should play a critical role in the design and operation. This paper describes such a system, the T-Sat1 nanosatellite. The satellite is being developed by undergraduate and graduate students at the University of Manitoba, as an extension of the standard classroom experience. Particular attention is given to the formation and maintenance of a team of teams, including the teams focusing on the satellite subsystems, assembly, integration and testing, as well as the teams of advisors from academia, aerospace industry, other industries, business, military, government, and other organizations such as the radio community. The paper describes the satellite characteristics, its mission, subsystems, as well as the development of specifications, protocols for verification, testing, launch and early operating procedures, as ell as concepts for nominal operations.
Sequence learning is one of the hard challenges to current machine learning technologies and deep neural network technologies. This paper presents a literature survey and analysis on a variety of neural networks towar...
详细信息
ISBN:
(纸本)9781538633618
Sequence learning is one of the hard challenges to current machine learning technologies and deep neural network technologies. This paper presents a literature survey and analysis on a variety of neural networks towards sequence learning. The conceptual models, methodologies, mathematical models and usages of classic neural networks and their learning capabilities are contrasted. Advantages and disadvantages of neural networks for sequence learning are formally analyzed. The state-of-the-art, theoretical problems and technical constraints of existing methodologies are reviewed. The needs for understanding temporal sequences by unsupervised or intensive-training-free learning theories and technologies are elaborated.
Knowledge bases are a fundamental platform for knowledge acquisition, retaining, retrieval, reasoning and generation across machine learning, natural language processing and computational intelligence. The mathematica...
详细信息
ISBN:
(纸本)9781538633618
Knowledge bases are a fundamental platform for knowledge acquisition, retaining, retrieval, reasoning and generation across machine learning, natural language processing and computational intelligence. The mathematical model of formal concepts is centric in knowledge bases for modeling the basic unit of human knowledge and thinking threads. This paper presents the design and implementation of a cognitive knowledge base. The structure of the knowledge base is created as a dynamic concept network mimicking human knowledge represented in the brain. The knowledge base enables a set of novel knowledge manipulations for machine learning such as knowledge acquisition, access, analysis, refinement, fusion and system maintenance. Experimental results demonstrate the performance and efficiency of the implementation of the generic knowledge base for cognitive robots and machine learning systems.
To analyse the DeepFakes, which are AI-generated synthetic media that impersonate real people and pose substantial threats to digital content security, privacy, and authenticity. This research investigates a unique te...
详细信息
ISBN:
(数字)9798350389609
ISBN:
(纸本)9798350389616
To analyse the DeepFakes, which are AI-generated synthetic media that impersonate real people and pose substantial threats to digital content security, privacy, and authenticity. This research investigates a unique technique to DeepFake detection that employs Recurrent neural networks (RNN) are used for temporal sequence analysis and convolutional neural networks (CNN) for extracting spatial features. The suggested method seeks to improve detection accuracy across a wide range of DeepFake datasets by combining the strengths of CNNs and RNNs. The experimental results show that this hybrid architecture is excellent at identifying DeepFakes with good precision and recall.)
Comprehension is the action or capability of understanding. Comprehension is a higher cognitive process of the brain that searches relations between a given object or attribute and other objects, attributes, and relat...
详细信息
ISBN:
(纸本)0769519865
Comprehension is the action or capability of understanding. Comprehension is a higher cognitive process of the brain that searches relations between a given object or attribute and other objects, attributes, and relations in the long-term memory, and establishes a representational model for the object or attribute by connecting it to appropriate clusters of memory. It is recognized that although knowledge and information are powerful, before any information can be possessed and processed, it should be comprehended properly. This paper intends to model one of the cognitive life functions, comprehension, in cognitive informatics. The mechanism and process of comprehension are described based on the cognitive model of the brain developed by Wang. Then, real-time process algebra (RTPA) is used to formally describe the comprehension process.
Summary form only given. Massive data sets produced by smart cities, science and health systems throw up challenges of storing, analyzing and visualizing data for the purposes of benefitting society. We look at how ad...
详细信息
Summary form only given. Massive data sets produced by smart cities, science and health systems throw up challenges of storing, analyzing and visualizing data for the purposes of benefitting society. We look at how advances in knowledge representation, brain information processing and knowledge manipulation can enable health professionals to organize, visualize and draw inferences from patient-specific data. Case examples from medical images from MRI and CT scans will be used to illustrate the challenges of detecting small pathological features. The presentation will also focus on data analysis, problem solving, decision making and the need for active engagement of health professionals in the analytical process.
暂无评论