scholarly journals Integrating Manual and Automatic Annotation for the Creation of Discourse Network Data Sets

2020 ◽  
Vol 8 (2) ◽  
pp. 326-339 ◽  
Author(s):  
Sebastian Haunss ◽  
Jonas Kuhn ◽  
Sebastian Padó ◽  
Andre Blessing ◽  
Nico Blokker ◽  
...  

This article investigates the integration of machine learning in the political claim annotation workflow with the goal to partially automate the annotation and analysis of large text corpora. It introduces the MARDY annotation environment and presents results from an experiment in which the annotation quality of annotators with and without machine learning based annotation support is compared. The design and setting aim to measure and evaluate: a) annotation speed; b) annotation quality; and c) applicability to the use case of discourse network generation. While the results indicate only slight increases in terms of annotation speed, the authors find a moderate boost in annotation quality. Additionally, with the help of manual annotation of the actors and filtering out of the false positives, the machine learning based annotation suggestions allow the authors to fully recover the core network of the discourse as extracted from the articles annotated during the experiment. This is due to the redundancy which is naturally present in the annotated texts. Thus, assuming a research focus not on the complete network but the network core, an AI-based annotation can provide reliable information about discourse networks with much less human intervention than compared to the traditional manual approach.

2020 ◽  
Vol 0 (0) ◽  
Author(s):  
Bakhe Nleya ◽  
Philani Khumalo ◽  
Andrew Mutsvangwa

AbstractHeterogeneous IoT-enabled networks generally accommodate both jitter tolerant and intolerant traffic. Optical Burst Switched (OBS) backbone networks handle the resultant volumes of such traffic by transmitting it in huge size chunks called bursts. Because of the lack of or limited buffering capabilities within the core network, burst contentions may frequently occur and thus affect overall supportable quality of service (QoS). Burst contention(s) in the core network is generally characterized by frequent burst losses as well as differential delays especially when traffic levels surge. Burst contention can be resolved in the core network by way of partial buffering using fiber delay lines (FDLs), wavelength conversion using wavelength converters (WCs) or deflection routing. In this paper, we assume that burst contention is resolved by way of deflecting contending bursts to other less congested paths even though this may lead to differential delays incurred by bursts as they traverse the network. This will contribute to undesirable jitter that may ultimately compromise overall QoS. Noting that jitter is mostly caused by deflection routing which itself is a result of poor wavelength and routing assigning, the paper proposes a controlled deflection routing (CDR) and wavelength assignment based scheme that allows the deflection of bursts to alternate paths only after controller buffer preset thresholds are surpassed. In this way, bursts (or burst fragments) intended for a common destination are always most likely to be routed on the same or least cost path end-to-end. We describe the scheme as well as compare its performance to other existing approaches. Overall, both analytical and simulation results show that the proposed scheme does lower both congestion (on deflection routes) as well as jitter, thus also improving throughput as well as avoiding congestion on deflection paths.


Author(s):  
Peyakunta Bhargavi ◽  
Singaraju Jyothi

The moment we live in today demands the convergence of the cloud computing, fog computing, machine learning, and IoT to explore new technological solutions. Fog computing is an emerging architecture intended for alleviating the network burdens at the cloud and the core network by moving resource-intensive functionalities such as computation, communication, storage, and analytics closer to the end users. Machine learning is a subfield of computer science and is a type of artificial intelligence (AI) that provides machines with the ability to learn without explicit programming. IoT has the ability to make decisions and take actions autonomously based on algorithmic sensing to acquire sensor data. These embedded capabilities will range across the entire spectrum of algorithmic approaches that is associated with machine learning. Here the authors explore how machine learning methods have been used to deploy the object detection, text detection in an image, and incorporated for better fulfillment of requirements in fog computing.


Author(s):  
Antonios Danalis

The popularity of the World Wide Web has led to an exponential increase of the traffic generated by its users for over a decade. Such a growth, over such a long period of time, would have saturated both the content providers and the network links had Web caching not been efficiently deployed. Web caching can improve the overall performance of the World Wide Web in several ways, depending on the decisions made regarding the deployment of the corresponding caches. By placing caches in strategic positions, the core network traffic can be reduced, the load of a content provider can be scaled down, and the quality of service, as the users perceive it, can be improved. In this article we present an overview of the major design and implementation challenges in Web caching, as well as their solutions.


2003 ◽  
Vol 12 (02) ◽  
pp. 241-273 ◽  
Author(s):  
ANA L. C. BAZZAN ◽  
ROGÉRIO DUARTE ◽  
ABNER N. PITINGA ◽  
LUCIANA F. SCHROEDER ◽  
FARLON DE A. SOUTO ◽  
...  

This work reports on the ATUCG environment (Agent-based environmenT for aUtomatiC annotation of Genomes). It consists of three layers, each having several agents in charge of performing repetitive and time-consuming tasks. Layer I aims at automating the tasks behind the process of finding ORFs (Open Reading Frames). Layer II (the core of our approach) is associated with three main tasks: extraction and formatting of data, automatic annotation of data regarding profiles or families of proteins, and generation and validation of rules to automatically annotate the Keywords field in the SWISS-PROT database. Layer III permits the user to check the correctness of the automatic annotation. This environment is being designed having the sequencing of the Mycoplasma hyopneumoniae in mind. Thus examples are presented using data of organisms of the Mycoplasmataceae family. We have concentrated the developments in layer II because this is the most general one and because it focusses on machine learning algorithms, a characteristic which is not usual in annotation systems. Results regarding this layer show that with learning (individual or colaborative), agents are able to generate rules for annotation which achieve better results than those reported in the literature.


Author(s):  
Vasilis Friderikos ◽  
Giorgos Chochlidakis ◽  
Hamid Aghvami ◽  
Mischa Dohler

The 5th Generation wireless and mobile communication is expected to provide ultrahigh data rates over wireless in the range of Gbps. But 5G will also be about providing consistency and supporting Quality of Experience in a personalized manner. We foresee an evolution in terms of physical layer enhancements to provide increased data rates, whereas a revolutionary step is required in terms of network orchestration and management to provide consistency and efficient utilization of the available resources at a minimum cost. In this chapter, key trends in wireless access technologies and thus-required network management strategies with respect to the core network are discussed. In the roadmap towards 5G networks, we envision an evolution of technologies for supporting Gbps wireless transmission, whereas a revolution would be required from the current modus operandi in the ways network orchestration and resource management is performed in these complex, hierarchical, heterogeneous and highly autonomous wireless networks.


Author(s):  
Antonios Danalis

The popularity of the World Wide Web has led to an exponential increase of the traffic generated by its users for over a decade. Such a growth, over such a long period of time, would have saturated both the content providers and the network links had Web caching not been efficiently deployed. Web caching can improve the overall performance of the World Wide Web in several ways, depending on the decisions made regarding the deployment of the corresponding caches. By placing caches in strategic positions, the core network traffic can be reduced, the load of a content provider can be scaled down, and the quality of service, as the users perceive it, can be improved. In this article we present an overview of the major design and implementation challenges in Web caching, as well as their solutions.


Author(s):  
Kashinath Basu

A significant proportion of the traffic on the 4th generation of mobile networks (4G) will be interactive multimedia traffic. This chapter presents the development and evaluation of an edge device model for the lu interface of a 4G network for mapping the Quality of Service (QoS) requirements and traffic characteristics of aggregated IP traffic flows belonging to multiple classes of continuous media (Audio and Video) sources and data classes from the core network onto a single ATM Virtual Channel (VC) at the access network. This model was developed as part of a wider range of research activity focused on supporting QoS in future mobile networks.


2017 ◽  
Vol 7 (1) ◽  
pp. 36-40 ◽  
Author(s):  
Joana Pereira ◽  
Hugo Peixoto ◽  
José Machado ◽  
António Abelha

Abstract The large amounts of data generated by healthcare transactions are too complex and voluminous to be processed and analysed by traditional methods. Data mining can improve decision-making by discovering patterns and trends in large amounts of complex data. In the healthcare industry specifically, data mining can be used to decrease costs by increasing efficiency, improve patient quality of life, and perhaps most importantly, save the lives of more patients. The main goal of this project is to apply data mining techniques in order to make possible the prediction of the degree of disability that patients will present when they leave hospitalization. The clinical data that will compose the data set was obtained from one single hospital and contains information about patients who were hospitalized in Cardio Vascular Disease’s (CVD) unit in 2016 for having suffered a cardiovascular accident. To develop this project, it will be used the Waikato Environment for Knowledge Analysis (WEKA) machine learning Workbench since this one allows users to quickly try out and compare different machine learning methods on new data sets


Author(s):  
Bharathi N. Gopalsamy ◽  
Brindha G. R. ◽  
B. Santhi

Machine learning (ML) is prevalent across the globe and applied in almost all domains. This chapter focuses on implementation of ML with real-time use cases. Day-to-day activities are automated to ease the task and increase the quality of decision. ML is the backbone of the perfect decision support system with a plethora of applications. The use case described in this chapter is ML & Security, which is implemented in R Script. Adversaries took advantages of ML to avoid detection and evade defenses. Network intrusion detection system (IDS) is the major issue nowadays. Its primary task is to collect relevant features from the computer network. These selected features can be fed into the ML algorithms to predict the label. The challenge in this use case is what type of feature to consider for intrusion and anomaly detection (AD). This chapter focuses on end-to-end process to get insight into the stream of data from the network connection with priority given to forecasting mechanism and prediction of the future. Forecasting is applied to the time series data to get sensible decisions.


Author(s):  
Igor Bisio ◽  
Stefano Delucchi ◽  
Fabio Lavagetto ◽  
Mario Marchese ◽  
Giancarlo Portomauro ◽  
...  

The main contribution of this chapter is the description of HySEP, Hybrid Simulated-Emulated Platform, developed by the authors and aimed at simulating/emulating heterogeneous networks to validate and test algorithms for traffic control and Quality of Service (QoS) assurance. Main features of HySEP are the appropriate level of accuracy and detail reached by using equipments available in most communication research laboratories, at low cost, and the easy configurability. HySEP is divided into three parts connected each others: the emulated core network; the simulated wireless access network communicating with the core network; and the real remote host. The overall platform is able to handle real traffic flows and overcomes the limitations introduced by other network simulators. HySEP is characterized by remarkable versatility and wide applicability to support the validation of different algorithms.


Sign in / Sign up

Export Citation Format

Share Document