This paper discusses the use of federated learning as a method for optimizing decision-making in communication systems. Federated learning is a machine learning technique that enables the training of models on decentr...
详细信息
The software company FERK-Systems has been providing mobile health care information systems for various German medical services (e.g. Red Cross) for many years. Since handwriting is an issue in the medical and health ...
详细信息
ISBN:
(纸本)9789898425706
The software company FERK-Systems has been providing mobile health care information systems for various German medical services (e.g. Red Cross) for many years. Since handwriting is an issue in the medical and health care domain, a system for handwriting recognition on mobile devices has been developed within the last few years. While we have been continually improving the degree of recognition within the system, there are still changes necessary to ensure the reliability that is imperative in this critical domain. In this paper, we present the major improvements made since our presentation at the ICE-B 2010, along with a recent real-life usability evaluation. Moreover, we discuss some of the advantages and disadvantages of current systems, along with some business aspects of the vast, and growing, mobile handwriting recognition market.
We consider the min-max graph balancing problem with strict negative correlation (SNC) constraints. The graph balancing problem arises as an equivalent formulation of the classic unrelated machine scheduling problem, ...
详细信息
Networks-on-Chips (NoCs) are meeting the growing inter-tile communication needs of multicore chips. However, achieving system scalability by utilizing hundreds of cores on-chip requires high performance, yet energy-ef...
详细信息
ISBN:
(纸本)9781479918713
Networks-on-Chips (NoCs) are meeting the growing inter-tile communication needs of multicore chips. However, achieving system scalability by utilizing hundreds of cores on-chip requires high performance, yet energy-efficient on-chip interconnects. As electrical interconnects are marred by high energy-to-bandwidth costs, threatening multicore scalability, on-chip nanophotonics, which offer high throughput, yet energy-efficient communication, are an alternative attractive solution. In this paper we consider silicon nanophotonic components that are embedded completely within the silica (SiO 2 ) substrate as opposed to prior-art that utilizes die on-surface silicon nanophotonics. As nanophotonic components now reside in the silica substrate's subsurface, a greater portion of a chip's real estate can be utilized by cores and routers, while non-obstructive interconnect geometries offering higher network throughput can be implemented. First, we show using detailed simulations based on commercial tools that such silicon-in-silica (SiS) structures are feasible, and then demonstrate our proof of concept by utilizing a hybrid SiS-based photonic mesh-diagonal links topology that provides both higher effective throughput and throughput-to-power ratio versus prior-art.
A multiferroic tunnel junction (MFTJ) is a promising device for future memory systems with discrete and different logic states which are controlled by a combination of electric and magnetic fields. The goal of ongoing...
详细信息
Reliability, efficiency and continuity of power energy supplied is an area which receives increasing attention as the main infrastructure of power transmission and distribution systems in many countries is ageing. Hot...
Reliability, efficiency and continuity of power energy supplied is an area which receives increasing attention as the main infrastructure of power transmission and distribution systems in many countries is ageing. Hot spot related cable failures and power interruptions have a big financial impact on the power provider. The operational parameters of a method to identify in real time hot spots on cables is investigated in this paper, which can result to timely fault prevention. This can be achieved through exploiting the optical sensing capabilities of the existing optical fibre on-grid network. The resulting parameters aim to be advantageous for the system as they increase accuracy and will enable the Remote, automated, continuous, and real time monitoring of the grid infrastructure integrity.
The Internet of Things is a collection of devices that communicate by exchanging a variety of data among them, in which time synchronization is needed for meaningful information creation and transmission. The robustne...
详细信息
Many insurance companies today deal with the issue of fraudulent insurance claims, which results in significant yearly financial loss. Since the losses are covered by raising policyholders’ premium costs, these fraud...
详细信息
ISBN:
(纸本)9781665475778
Many insurance companies today deal with the issue of fraudulent insurance claims, which results in significant yearly financial loss. Since the losses are covered by raising policyholders’ premium costs, these frauds have a negative impact on society. The traditional claim investigation procedure has also been blamed for producing unreliable conclusions because it is time-consuming and laborious. Therefore, using machine learning and the XGBoost method, we construct an automated fraud detection application framework in this study. Accurately identifying fraud claims in a shorter amount of time is the goal. Data analysis is utilized throughout the process to validate, sanitize, and extract the pertinent data. As a result, the insurance firm can retain its reputation outside by employing this structure and has a reliable relationship with clients that they can share.
The paper demonstrates the improvement in Influenza A classification based on viral host when applying feature selection on classical machine learning techniques. The impact of using the most informative DNA positions...
详细信息
ISBN:
(纸本)9789881925275
The paper demonstrates the improvement in Influenza A classification based on viral host when applying feature selection on classical machine learning techniques. The impact of using the most informative DNA positions on classifier efficiency and performance was measured. Both decision trees (DTs) and neural networks (NNs) were used. The experiments were conducted on DNA sequences belonging to the PB1 and HA segments of subtypes H1 and H5 respectively. Sequences from each segment were further divided into human and nonhuman hosts prior to classification analysis. Accuracy, sensitivity, specificity, precision and time were used as performance measures. Extracting the best hundred informative positions with information gain increased classification efficiency by 90% for both classifiers, without compromising performance significantly. NNs performed better on both DNA segments than DTs, when decreasing the number of informative positions below a hundred. The classification speed of NNs was improved vastly compared to DTs, when classifying the H1, PB1 segment.
There are quintillions of data on deoxyribonucleic acid(DNA)and protein in publicly accessible data banks,and that number is expanding at an exponential *** scientific fields,such as bioinformatics and drug discovery,...
详细信息
There are quintillions of data on deoxyribonucleic acid(DNA)and protein in publicly accessible data banks,and that number is expanding at an exponential *** scientific fields,such as bioinformatics and drug discovery,rely on such data;nevertheless,gathering and extracting data from these resources is a tough *** data should go through several processes,including mining,data processing,analysis,and *** study proposes software that extracts data from big data repositories automatically and with the particular ability to repeat data extraction phases as many times as needed without human *** software simulates the extraction of data from web-based(point-and-click)resources or graphical user interfaces that cannot be accessed using command-line *** software was evaluated by creating a novel database of 34 parameters for 1360 physicochemical properties of antimicrobial peptides(AMP)sequences(46240 hits)from various MARVIN software panels,which can be later utilized to develop novel ***,for machine learning research,the program was validated by extracting 10,000 protein tertiary structures from the Protein Data *** a result,data collection from the web will become faster and less expensive,with no need for manual data *** software is critical as a first step to preparing large datasets for subsequent stages of analysis,such as those using machine and deep-learning applications.
暂无评论