The current paper is a part of a master dissertation carried out at the Open University of Cyprus. Starting from an individual to a multinational corporation level we are looking to find, understand and define the purpose, the need and the impact of managing the impressions made. We argue that strategies may vary regarding the type of business. We suggest that anybody has an opinion as soon as they identify the brand. This opinion is disseminated and collectively can be seen as reputation. The progress in communication is implying bigger social influence between the people, trends are changing fast. We are presenting different tools to find these insights, which perform Online Reputation Management (ORM), free, paid or open source. Using theory from social science and technological facts, we discuss possible caveats. Finally we aim to use python scripts in order to collect, preprocess and analyze real data to see it in practice through the prism of a real business entity.
The increasing importance of wireless communications, involving electronic devices, has been widely recognized. Performance is a fundamental issue, resulting in more reliable and efficient communications. Security is also crucially important. Laboratory measurements are presented about several performance aspects of Wi-Fi IEEE 802.11ac WPA2 point-to-point links. Our study contributes to performance evaluation of this technology under WPA2 encryption, using available equipments (Cisco 2702i access points and Edimax AC 1200 adapters). New results are given from TCP and UDP experiments concerning TCP throughput versus TCP packet length, jitter and percentage datagram loss versus UDP datagram size. Comparisons are made to results obtained for corresponding IEEE 802.11n links. Conclusions are drawn about the comparative performance of the links.
Dealing with data transmission in smart grid networks is one of the most challenging topics in telecommunications industry. Wireless mesh network (WMN) and Power line communication (PLC) are suitable technology for distributed smart grid communication; both connections are lossy connections. Network virtualization (NV) is a promising innovation to bolster customized end-to-end execution of different services in real-time. In this paper, a NV-based structure is proposed for smart grid communication, where real-time services are upheld by virtual networks that are mapped to both physical systems at the same time. Since the VN mapping and subcarrier task issue is nondeterministic polynomial-time hard, heuristic solutions are implemented. To reduce the complexity of the system, data sparsification algorithm is applied to the input of the system. The results with the data sparsification and the results without the data sparsification are compared in the proposed work. Simulation results show the adequacy of our proposed technique.
In this article, we propose a new approach for the churn analysis. Our target sector is Telecom industry, because most of the companies in the sector want to know which of the customers want to cancel the contract in the near future. Thus, they can propose new offers to the customers to convince them to continue using services from same company. For this purpose, churn analysis is getting more important. We analyze well-known machine learning methods that are logistic regression, Naive Bayes, support vector machines, artificial neural networks and propose new prediction method. Our analysis consist of two parts which are success of predictions and speed measurements. Affect of the dimension reduction is also measured for the analysis. In addition, we test our new method with a second dataset. Artificial neural networks is the most successful as we expected but our new approach is better than artificial neural networks when we try it with data set 2. For both data sets, new method gives the better result than logistic regression and Naive Bayes.
The improvement in the conventional optical Transport network (OTN) is exceptionally moderate regarding the rapid growth in the mobile and the IP-Core technologies, particularly with the requirements of the new 5G advances. There are many challenges in the OTN such the expensive cost of the multilayer services planning, the quality of the services and the quality of the resilience must be recovered first to cope with the changes in the new generations of the access communication networks. The needs to overcome many of these challenges become vital nowadays, and depend on many factors in the OTN such the status of the optical fiber cables, the flexibility, the responsive and the availability of OTN assets to the direct customer control. In this paper for the first time a new proposed model is introduced by reorganizing the OTN to fit the needs of the new generations of the communications market, the model consolidates two promising technologies with each other which are the Software Defined Network (SDN) and the Machine Learning (ML) to overcome the previous challenges and to reconstruct the traditional OTN to be more smart, virtualized and automated. The fundamental role of the SDN is to transform the services on the OTN to be more dynamic rather fixed, at the same time the aim of the ML such the Artificial Neural Network (ANN) is to help the centralized controller of the SDN in the OTN by the past experience of the performance of the optical links in the OTN, this enables the centralized controller to formulate the right decisions about the optimized routes of the services restoration between the different domains and multilayers in the OTN. For the first time, the optical cloud concepts are introduced in the OTN by slicing and virtualizing the various domains with its vendors in the heterogeneous optical network to 3 integrated unified layers, this provides the required resiliencies and the bandwidth on demand between the multilayers and the different domains in the OTN in a more elastic way. The model is tested using 2 methods , the 1'st method is done by a software simulation by using a SPSS software model and its input data was 500 records from real OTN , and the 2'nd method was done by performing practical case study on the long distances heterogeneous OTN network in one of the middle east countries about the integrations between the different optical network domains, slicing the optical network, and the centralized controller to reconstruct the heterogonous OTN to 3 layers to perform the resilience of the services of the multi failure in the same domain through the multilayers in optical network. The results of the new model according to the practical case study in the long-distance heterogamous OTN show that: The dependence on the single vendor is nearly neglected with applying the concept of the clouding and slicing in the heterogeneous OTN, the pay for the end-users bandwidths has become possible and the time to provide the bandwidth on demand has become very short , the meshing between the heterogeneous optical network became available and the resilience for diamond services improved from 25% for double or triple faults to 100% after applying part of our model in the long distance optical network, the available bandwidth of the optical core network in the long distance network is optimized by more than 25% , the revenue from some OTN domains which have free bandwidths more than 50 % is increased by more than 50%, the switching time enhanced by about 50%, and the latency reduced from 27 msec to 742 usec for the selected routes which is optimized from the centralized controller.
