Internet Subscribers Research Articles

Willingness to pay for broadband: A case study of Wisconsin

As broadband expansion efforts in the U.S. continue with historic investments, consumer demand for residential broadband services is of first-order importance. Several past broadband willingness to pay studies estimate the value of broadband to be low, compared to the current national average cost of internet subscriptions being around $65 per month. Additional estimates of household's willingness to pay for broadband are needed to provide more insight into how much households value these services and what type of support would be most efficient in increasing broadband adoption to close the digital divide. These estimates are helpful for policymakers, internet service providers, and community leaders working on broadband subsidy programs, trying to bring ISPs to communities, and setting service prices. Using choice experiment data from a 2022 broadband survey in Wisconsin, I find households are willing to pay $45-$334 per month to go from not having internet at home to having speeds between 25 Mbps and 1,000 Mbps and are willing to pay $10-$38 for greater reliability, depending on household income level. While consumers are willing to pay more for speed than reliability, the value consumers place on speed is diminishing. Connection is valuable to people, but getting connected at a much faster speed is not seen as worth a lot more. These estimates suggest households place a high value on home internet and are willing to pay at least, if not more, than prices already available on the market in many places.

Examining the Influence of ICT Penetration on Gender Disparities: A Data Panel Econometric Analysis of the Seven Southeast Asian Developing Countries

The aim of the study is to analyze to which Information and Communications Technology (ICT) penetration affects gender disparities in seven Southeast Asia’s developing countries. ICT indicators are measured through the individuals using the internet, mobile phone subscriptions, and internet subscriptions. Aside, the seven Southeast Asia’s developing countries include Indonesia, Cambodia, Laos, Malaysia, Thailand, Philippines, and Vietnam. These countries are selected as they have comparable economic development stage from lower to upper-middle income economics on the basis 2022’s World Bank Country Income Classification. The data panel econometrics is carried out to measure the effect of the independent variables towards gender disparity. Gender disparity is represented by the Gender Inequality Index. The time series data is yearly from 2000 to 2022. The econometric assessment shows the individual internet ownership and subscription have a negative and significant effect to gender disparity. It means that more internet penetration can lower the gender disparity in those seven countries. The academic implication is to enhance economic consideration towards more internet penetration and access to reduce gender disparity. In the heart of policymaking, our research indicates more internet accessibility to reduce gender disparity. As in the common knowledge, the internet is a source of information and knowledge to society

Analyzing the impact of cultural accessibility and ICT infrastructure on economic growth in Kazakhstan

For this, the primary aim of this study was to analyze of the impact of cultural accessibility and ICT (information and communication technology) infrastructure on economic growth in Kazakhstan, employing regression models to asses a single country data from 2008 to 2022. The research focuses on two sets of variables: cultural development variables (e.g., number of theaters, museums, and others) and ICT infrastructure variables (e.g., number of fixed Internet subscribers, total costs of ICT, and others). Principal component analysis (PCA) as employed to reduce the dimensionality of the data and identify the most significant predictors for the regression models. The findings indicate that in the cultural development model (Model 1), the number of recreational parks and students are significant positive predictors of GDP per capita. In the ICT infrastructure model (Model 2), ICT costs are found to have a significant positive impact on GDP per capita. Conversely, traditional connectivity indicators, such as the number of fixed telephone lines, show a low dependence on economic growth, suggesting diminishing returns on investment in these outdated forms of ICT. These results suggest that investments in cultural and ICT infrastructure are crucial for economic development. The study provides valuable insights for policymakers, emphasizing the need for quality improvements in education and strategic modernization of communication technologies.

Exploring Patterns and Disparities in E-Consult Referrals: An Analysis of Patient and Community Factors in Colorado Health Care.

Objective: Electronic consultations (e-consults) provide a strategic solution to address challenges in health care systems related to cost management and access to care. This study aims to investigate the multilevel patient characteristics associated with higher frequency of receiving e-consults and increased likelihood of completion. Materials and Methods: University of Colorado's electronic medical record were analyzed to study factors influencing referral types (e-consult vs. standard) and their completion rates from April 2018 to September 2023. Multivariate probit regression assessed the impact of patient-level and community-level factors (urban-rural classification, Social Vulnerability Index, and technology accessibility) on e-consult referrals and completion. Results: In 263,882 records, 92.5% were standard referrals, and 7.4% were e-consult referrals. Analysis showed that females were less likely than males (OR = 0.95, 95%CI[0.93, 0.96]), and Blacks were more likely than Whites (OR = 1.03, 95%CI[1.01,1.06]) to receive e-consult referrals. Medicaid patients had lower odds compared to those with Medicare only (OR = 1.04, 95%CI[1.00,1.07]), and rural residency was associated with lower odds (OR = 0.80, 95%CI[0.73,0.88]) of e-consult referral. Factors such as areas with higher population without internet subscription (OR = 1.03, 95%CI[1.01,1.04]) and higher social vulnerabilities (OR = 1.26, 95%CI[1.16,1.37]) increased e-consult odds. Black patients were less likely to have their referrals completed compared to Whites. Patients who resided in regions with limited computer and smartphone access, as well as higher social vulnerabilities, showed decreased odds of referral completion. Discussions and Conclusion: This study highlights the need for partnering with a variety of health care organizations, especially those serving low-income and disadvantaged populations, to enhance health care access equity through the use of e-consults.

Embracing the digital revolution: Examining the relationship between ICT adoption and carbon emissions in the Persian Gulf.

In this digital age, promoting economic development through technology innovation and adoption has become a pressing matter, contributing to increased productivity and, in turn, carbon emissions. Consequently, this study employs a novel technique (Newey-West Standard Error Method, Technology Adaptation Model) to quantify information and communication technology (ICT) adoption rates as a proxy indicator for evaluating the Persian Gulf economy's technology development. Moreover, this study investigates the evidence of the environmental Kuznets curve, with trade openness, technological adoption, and innovation as sustainable development controls. The findings reveal that two of three technological innovation instruments, fixed telephone, and internet subscriptions, increase carbon emissions. In contrast, mobile cellular subscriptions simultaneously reduce carbon emissions in the Persian Gulf. Furthermore, measures of technology adoption, high-technology exports, and electricity use contribute to the increase in carbon emissions. Trade openness also raises carbon emissions in the Persian Gulf. These findings suggest that policymakers must develop technological innovation and adoption strategies that effectively promote a greener environment.

The Economic Efficiency of Android Phone Enabled Modems and Standard Modems in Nigeria: The Case of Abuja

This study examined the economic efficiency of Android phone-enabled modems and standard modems in Nigeria. A sample of 388 commercial (cybercafe) and private internet service users were selected using random sampling. The data were analyzed using frequency and percentage calculations to identify patterns and trends as well as ordinal logit regression analysis. The research study revealed that the odds for higher economic efficiency rise when subscribers or commercial (cybercafe) and private internet service users’ monthly internet subscriptions is lower. Similarly, the study revealed that the odds for higher economic efficiency rise when subscribers or commercial (cybercafe) and private internet service user’s daily internet access time increases. Regarding connectivity issues while using an Android-enabled modem, the study showed that the odds for higher economic efficiency fall when connectivity challenges increase. The odds for higher economic efficiency increase when connectivity challenges arise while using a standard modem. The research study also showed that the odds for higher economic efficiency increases when more expensive Android phones are used. The study further revealed that the odds for higher economic efficiency will fall when subscribers or commercial (cybercafe) and private internet service users use less expensive standard modems. The study also revealed that the effect of the explanatory variables or predictors is different across the levels of the dependent or explained variable. Thus, the study recommends that regulatory agencies and operators should come up with a price per megabyte of internet access that will reduce the initial cost of services rendered by private and commercial internet users in a bid to drive economic growth. Regulators and service providers must ensure minimal or no interference with the service provided to end users in a bid to promote the economic efficiency of Internet services. Efforts must be made by the government to the efficient devices affordable, and an embargo must be placed on the importation of internet devices (phone and standard modem) with poor efficiency

Application of Zoom E-learning Management System in Business Education in Universities Nigeria

The study examined the application of the Zoom e-learning management system, in the context of Business Education in Nigerian universities. The study employed a descriptive survey design. The sample consisted of 801 lecturers. A questionnaire consisting of 39 items was utilized for data gathering. The instrument demonstrated face and content validity, with reliability indexes of 0.86, 0.91, and 0.84, respectively. The data analysis was conducted using the measures of statistical mean. The results indicated that the majority of participants expressed their preference for the Zoom management system as the primary method of instruction in university business education in this era of ICT worldwide dominance. Several obstacles have been identified that have impeded the complete integration of Zoom into teaching and learning in Nigerian universities. The challenges encompass insufficient infrastructure for Zoom utilization, exorbitant subscription fees, and insufficient digital proficiency among teachers and students. Consequently, the study recommended the necessity of the government for enhanced financing to provide the necessary equipment and readily available internet subscriptions in universities. School authorities are anticipated to establish the integration of the Zoom platform as a mandatory policy.

Smart City Projects Financing

The implementation of advanced digital technologies and innovations in smart cities for the provision of more efficient, sustainable, and inclusive city services, the development of infrastructure, and the improvement of citizens’ quality of life require appropriate financing technologies. In addition to purely public (for example, green bonds, social impact bonds, state subsidies, etc.) and purely private (venture investments, leasing, etc.) financial instruments, various options for public-private partnerships and financing with and against the participation of the community (participatory ), which increase the efficiency of budget financing due to the agreed distribution of risks and responsibilities between stakeholders. The bibliometric analysis of publications indexed by Scopus using the keywords “smart city” and “public-private partnership” using the VOSviwer tool allowed establishing the periods of growth (2013-2019 and 2022-2023) and decline (2019-2022) of the popularity of this topic among scientists, countries-leaders of research activity (USA, India, China, Italy, Spain, Great Britain, Canada and Germany), dominant directions of cross-sectoral research. The analysis of the Smart City Index Report (IMD Smart City Index Report) showed that in 2023, the top 5 leaders were Switzerland (Zurich), Norway (Oslo), Australia (Canberra), Denmark (Copenhagen), and Great Britain (London). With the help of the Google Trends toolkit, an analysis of the dynamics of Internet requests over the past 10 years from citizens of these countries, as well as Ukraine and the world, was carried out using identical keywords. The analysis proved that both in these countries and in the world as a whole, public interest in the development of the smart grid was recorded (in Ukraine, the highest peak (100 GT Scale) was in 2019; in Great Britain and Switzerland – in 2018, in Denmark – in 2017 and 2022, in Australia and Norway – in 2016) against the background of an almost absolute (except for Great Britain and Australia) lack of public interest in issues of public-private partnership. A regression model was developed to study the impact of participatory financing on the readiness level of Ukraine’s regions (as of 2020) to implement smart local development technologies. An integral indicator was used as the resulting variable, within which data on the assessment of the availability of electronic public services and the level of automation and digitization of public services (the number of state and local self-government bodies that provided the possibility of using electronic democracy tools), the level of use electronic platforms for communication with state bodies (the number of registered “E-appeals”, published “E-petitions” and reports on “E-consultations”), the availability of the Internet (the number of Internet subscribers and the share of households that have access to Internet services home). In the modeling process, adjustments were made for the gross regional product and the population of the respective region, the weighting factors were determined by the method of principal components, the estimation of the model parameters was carried out by the method of least squares, the calculations were made using the MS Excel spreadsheet and the Statistica application program package.

Implementation of 5G Telecommunication Network Services in Indonesia based on Techno-economic Analysis

The 2300 MHz spectrum is a medium band that telco operators will not pay much attention to when they deploy 5G. They are more comfortable at 2.6 GHz, 3.5 GHz, 26 GHz, and 28 GHz, in addition to 700 MHz for the breadth of coverage. The performance of cellular telecommunications services based on 5G technology is possible for new operators, although it will be carried out as stand-alone services. This opportunity will be taken by looking at internet subscriber data/data communication from existing operators as active internet users, which is quite large and has a potential of over 250 million users. There has been no previous study regarding the feasibility of deploying this 5G technology-based Broadband Wireless Access (BWA) Network. Based on the experience of implementing previous generations of telecommunication service technology, the government and operators need to be careful in determining the right moment to deploy this 5G technology service, which is predicted to be able to provide broadband services with streaming capabilities of 10 to 100 times the streaming speed of 4G technology. It should be noted that the lack of success of 3G performances in 2006 from 2G, 2.5G, and 2.75 G. Almost all operators who were expected to be very lucky turned out to be not optimal; even now, only 4 operators are playing on 3G. where they have not been able to force users of the 2G generation to switch to 3G, including in big cities where the performance of the 3G network is not yet optimal and evenly distributed. Still, many areas are blank spots from 3G networks and services. From this experience, scientific studies are needed to ensure the feasibility of the upcoming 5G BWA business and identify business opportunities that can be implemented. The feasibility analysis must be viewed from various aspects, namely aspects of technical readiness, market aspects, and financial aspects in terms of the techno-economics of the operators who will provide 5G telecommunications services by calculating several essential parameters as a measure of business feasibility, namely Net Present Value (NPV), Internal Rate of Return (IRR), and Payback Period (PBP).

Implementation of 5G Telecommunication Network Services in Indonesia based on Techno-economic Analysis

The 2300 MHz spectrum is a medium band that telco operators will not pay much attention to when they deploy 5G. They are more comfortable at 2.6 GHz, 3.5 GHz, 26 GHz, and 28 GHz, in addition to 700 MHz for the breadth of coverage. The performance of cellular telecommunications services based on 5G technology is possible for new operators, although it will be carried out as stand-alone services. This opportunity will be taken by looking at internet subscriber data/data communication from existing operators as active internet users, which is quite large and has a potential of over 250 million users. There has been no previous study regarding the feasibility of deploying this 5G technology-based Broadband Wireless Access (BWA) Network. Based on the experience of implementing previous generations of telecommunication service technology, the government and operators need to be careful in determining the right moment to deploy this 5G technology service, which is predicted to be able to provide broadband services with streaming capabilities of 10 to 100 times the streaming speed of 4G technology. It should be noted that the lack of success of 3G performances in 2006 from 2G, 2.5G, and 2.75 G. Almost all operators who were expected to be very lucky turned out to be not optimal; even now, only 4 operators are playing on 3G. where they have not been able to force users of the 2G generation to switch to 3G, including in big cities where the performance of the 3G network is not yet optimal and evenly distributed. Still, many areas are blank spots from 3G networks and services. From this experience, scientific studies are needed to ensure the feasibility of the upcoming 5G BWA business and identify business opportunities that can be implemented. The feasibility analysis must be viewed from various aspects, namely aspects of technical readiness, market aspects, and financial aspects in terms of the techno-economics of the operators who will provide 5G telecommunications services by calculating several essential parameters as a measure of business feasibility, namely Net Present Value (NPV), Internal Rate of Return (IRR), and Payback Period (PBP).

Implementation of 5G Telecommunication Network Services in Indonesia based on Techno-economic Analysis

The 2300 MHz spectrum is a medium band that telco operators will not pay much attention to when they deploy 5G. They are more comfortable at 2.6 GHz, 3.5 GHz, 26 GHz, and 28 GHz, in addition to 700 MHz for the breadth of coverage. The performance of cellular telecommunications services based on 5G technology is possible for new operators, although it will be carried out as stand-alone services. This opportunity will be taken by looking at internet subscriber data/data communication from existing operators as active internet users, which is quite large and has a potential of over 250 million users. There has been no previous study regarding the feasibility of deploying this 5G technology-based Broadband Wireless Access (BWA) Network. Based on the experience of implementing previous generations of telecommunication service technology, the government and operators need to be careful in determining the right moment to deploy this 5G technology service, which is predicted to be able to provide broadband services with streaming capabilities of 10 to 100 times the streaming speed of 4G technology. It should be noted that the lack of success of 3G performances in 2006 from 2G, 2.5G, and 2.75 G. Almost all operators who were expected to be very lucky turned out to be not optimal; even now, only 4 operators are playing on 3G. where they have not been able to force users of the 2G generation to switch to 3G, including in big cities where the performance of the 3G network is not yet optimal and evenly distributed. Still, many areas are blank spots from 3G networks and services. From this experience, scientific studies are needed to ensure the feasibility of the upcoming 5G BWA business and identify business opportunities that can be implemented. The feasibility analysis must be viewed from various aspects, namely aspects of technical readiness, market aspects, and financial aspects in terms of the techno-economics of the operators who will provide 5G telecommunications services by calculating several essential parameters as a measure of business feasibility, namely Net Present Value (NPV), Internal Rate of Return (IRR), and Payback Period (PBP).

Customers Reactions on Service Delivery by Major Telecommunications Operators in Nigerian Newspapers

Data from the Nigeria Communications Commission (NCC) stated that MTN lost a record of 667,245 Internet subscribers in one month in 2019.Figure shows that MTN’s total Internet subscribers stood at 52.2 million in July 2019 while the number dropped to 51.6 million in August 2019. It was observed that customers’ reactions bordered on issues ranging from unsolicited calls and short message services (SMS), unsubstantiated deductions of call credit units, poor service connection, network failure, cumbersome SIM swap process among others. This paper, thus, examines two Nigerian national dailies-The Punch and The Nation newspapers coverage of customers’ reaction to mobile telecommunication network services from 2015 to 2017. Using the content analysis research method, the study is anchored on the Social Exchange theory. The findings revealed a significant pattern of more negative than positive reactions by subscribers within the years under study. The negative complaints were high in 2015, dropped significantly in 2016, and rose again in 2017. On this, MTN recorded 15 (or 32%); Airtel 13 (or 28%), Glo10 (or 21%), while Etisalat (now 9Mobile) was 9 (or 19%), as against Etisalat (9 Mobile), 9 (or 19%), Glo 4 (or 8%) and Airtel 3 (or 6%) positive reactions by the customers during the period. Study thus recommends that Nigerian newspapers should once in a while carryout a survey on quality of service delivery by GSM operators and use such stories on the front page of their respective newspapers.

Prospects and setbacks for migrating towards 5G wireless access in developing Bangladesh: A comparative study

AbstractAs 4G technology served as a foundation, the emergence of 5G is now underway, ushering in a new era of connectivity. With the growing demand for seamless internet experiences, particularly in the face of escalating internet subscribers, the networking domain must evolve. Developing nations strive to align with this dynamic landscape, necessitating an upgrade that integrates internet of things (IoT), natural language processing (NLP), and artificial intelligence (AI) with network infrastructure. By enhancing networking systems with lower latency and improved scalability, 5G addresses congestion issues. This is achieved by coupling mm‐wave technology with the 5G framework through the 3rd Generation Partnership Project, significantly amplifying channel bandwidth. However, the comprehensive analysis underscores challenges in 5G implementation, encompassing aspects like distance, orientation, non‐line‐of‐sight conditions, protocol utilization, and server positioning. Drawing from a dataset provided by the University of Minnesota that illuminates the limitations of 5G implementation in select US cities, this paper extends its focus to densely populated regions in developing countries of the sub‐continent. Employing a machine learning approach, the paper delves into the constraints of 5G deployment in such areas. Ultimately, this research aims to shed light on the intricacies of 5G's implementation challenges, contributing to the discourse on enhancing network infrastructure in the evolving landscape of global connectivity.

Evaluation of Undergraduates and Postgraduates’ Utilisation of Internet for Academic Purpose in STREAM Fields

This study examines undergraduates and postgraduates use of the internet for academic purposes in STREAM fields. A mixed method research which sampled 100 respondents selected through snowballing sampling technique. Four research questions and one hypothesis guided this study. Two instruments namely Internet Usage Questionnaire (IUQ) and semi-structured interview on challenges experienced in the utilisation of internet (IUIT) were employed for data collection. Standard deviation and independent sample t-test analysed the descriptive (quantitative) aspect of questions and hypothesis raised respectively, while thematic analysis was employed in the narrative (qualitative) section. The study found that mobile apps, software, and website were among the media through which students access academic resources on the internet. No significant difference was found in the level of usage of the internet for academic purposes between undergraduates and postgraduates in STREAM fields. Poor internet facility, high cost of internet subscription, wide search for relevant academic resources due to inaccessibility to paid resources and many ad pop-ups owing to unlicensed software were among the challenges experienced by the respondents. Furthermore, stakeholders are enjoined to make available fast and affordable internet to ease the concerns raised.

Financial structure and CO2 emissions in Asian high-polluted countries: Does digital infrastructure matter?

In this study, we examined the impact of digital infrastructure and financial structure on the carbon emissions in high-polluted economies of Asia for the period of 1991–2019. For empirical analysis, we have applied panel ARDL-PMG and the quantile regression approach. The results of the ARDL model provided some important information. Firstly, our results suggest that financial structure help reduces carbon emissions in high-polluted Asian economies in the long run. Secondly, an increase in internet subscriptions in high-polluted Asian economies improves the environmental quality. Thirdly, a rise in the population and economic development of these countries positively impacts CO2 emissions. In contrast, a rise in R&D has a significant effect on CO2 emissions in the long run. The quantile regression approach also complements our findings of the ARDL-PMG model. The results of the quantile regression confirm that an improvement in financial structure in high-polluted Asia reduces CO2 emissions. However, the estimates are the largest from 40th to 70th quantiles in almost all models. Similarly, digitalization in the country also reduces CO2 emissions, and the estimates of digital variables appeared to be significantly negative in all quantiles.

Optimizing Internet Subscription for MTN Data Plan in a Specified Period

Optimizing internet subscription is the ability of an individual to be able to make decision of subscribing to a data plan with maximum benefit and minimum cost. To determine the optimal MTN monthly and yearly data plan to subscribe, linear programming was proposed and implemented to determine the optimal MTN data plan with minimum cost and maximum validity 30 days for the monthly plan and 360 days for the yearly plan. This study has shown it is better to subscribe for one daily of 1gb data plan, two days plan for 2gb, one week plan for 6gb and three times one week plan of 1000 naira for 2gb for the monthly subscription and also one daily of 1gb data plan, two days plan for 2gb, one week plan for 6gb and fifty times one week plan of 1000 naira for 2gb for the monthly subscription. This study will assist individuals as well as organizations is making decision of the data plan to subscribe to at a minimized rate. Individuals or organizations can gain insights on how to improve subscription of data plan for a specified period of time and also keeping track of the maximum gigabyte that can be affordable.

The usage of social media for learning and teaching in the built environment: a perspective from students

Advancement in information and communication technologies has led to the emergence of Social Media (SM). SM platforms are used to facilitate information sharing and learning. Literature shows that SM platforms have been associated with effective learning and improved student academic performance. Despite the recorded successes, the use of SM platforms to support students’ learning in built environment programmes has not been fully investigated. To address this knowledge gap, the current study examines the barriers and drivers encouraging the use of SM technologies to support learning and teaching in built environment programmes based on the perspective of university students. The data for the study were collected using semi-structured interviews and questionnaires. The mean score (MS) was used to analyse the quantitative data, while the qualitative data were analysed using thematic analysis. The top-two drivers for using SM technologies are the need for collaboration among students and compatibility with a twenty-first-century education. The top-two barriers to using SM platforms are the cost of internet subscriptions and the quality of internet services available in a particular location. The result shows that students are willing to incorporate elements of online learning. The research contributes to the current body of knowledge by highlighting the drivers and barriers to using SM to support learning. Moreover, the outcome of this study may guide administrators in higher institutions and other education policymakers, specifically for developing countries, to make informed decisions regarding integrating SM technologies into learning and teaching.

We need it: How Digital Marketing Tactics Influence the Purchasing Behaviour of Nigerian Millenials

Businesses have come to the realization that customers of today especially the millenials are actively engaged over the internet and to give them place utility, they need to engage them digitally. The objective of this study was to determine the influence of selected digital marketing tactics and their influence on the purchasing behavior of Nigerian Millenials. Research was conducted using cross-sectional survey design with data collected using close ended questionnaires designed with 5-point Likert scale questions. The study’s population comprised active internet subscribers in FCT within the millennial age group, and a sample size of 400 was arrived at using Taro Yamani’s formula. Linear regression analysis was used to test all three hypotheses, and results showed that all two selected digital marketing tactics (social media marketing and search engine optimization) have positive influence on purchasing behavior of Nigerian Millenials. Thus, it was concluded that increased use of digital marketing tactics would positively affect the purchasing behavior of Nigerian Millenials.

Does Real Income and Population Matter in Evaluating Domestic Trade Performance? Evidence from Nigeria

Domestic trade has been a key sector in the Nigerian economy in terms of economic sustainability, following its performance and resiliency in both tranquil and turbulent periods. Using Autoregressive Distributed Lag (ARDL) model for the period 2012Q3 to 2022Q3, this study examines the impact of factors driving trade growth in Nigeria. The findings show that growth in the country’s population and per capita income are the major factors driving domestic trade growth in Nigeria in both short-run and long-run periods. Also, changes in exchange rates positively impact domestic trade growth in the long-run period; information technology, in terms of growth in numbers of internet subscriptions, as well as trade openness, do not have a significant impact on trade growth. We suggest that the government expand internet coverage across the country and intentionally deepen the awareness of growing trade globalisation and its benefits to citizens and the nation’s economic growth.

Elephant Flows Detection Using Deep Neural Network, Convolutional Neural Network, Long Short-Term Memory, and Autoencoder

Currently, the widespread of real-time applications such as VoIP and videos-based applications require more data rates and reduced latency to ensure better quality of service (QoS). A well-designed traffic classification mechanism plays a major role for good QoS provision and network security verification. Port-based approaches and deep packet inspection (DPI) techniques have been used to classify and analyze network traffic flows. However, none of these methods can cope with the rapid growth of network traffic due to the increasing number of Internet users and the growth of real-time applications. As a result, these methods lead to network congestion, resulting in packet loss, delay, and inadequate QoS delivery. Recently, a deep learning approach has been explored to address the time-consumption and impracticality gaps of the abovementioned methods and maintain existing and future traffics of real-time applications. The aim of this research is then to design a dynamic traffic classifier that can detect elephant flows to prevent network congestion. Thus, we are motivated to provide efficient bandwidth and fast transmission requirements to many Internet users using SDN capability and the potential of deep learning. Specifically, DNN, CNN, LSTM, and Deep autoencoder are used to build elephant detection models that achieve an average accuracy of 99.12%, 98.17%, and 98.78%, respectively. Deep autoencoder is also one of the promising algorithms that do not require human class labeler. It achieves an accuracy of 97.95% with a loss of 0.13. Since the loss value is closer to zero, the performance of the model is good. Therefore, the study has a great importance to Internet service providers, Internet subscribers, as well as for future researchers in this area.