Search results for: green space planning models

Authors: Shari S. C. Shang, Lynn S. L. Chiu

Abstract:

In the digital payment market, some consumers use only one payment wallet while many others play multi-homing with a variety of payment services. With the diffusion of new payment systems, we examined the determinants of the adoption of multi-homing behavior. This study aims to understand how a digital payment provider dynamically expands business touch points with cross-business strategies to enrich the digital ecosystem and avoid the trap of multi-homing in customers. By synthesizing platform ecosystem literature, we constructed a two-dimensional research framework with one determinant of user digital behavior from offline to online intentions and the other determinant of digital payment touch points from convenient accessibility to cross-business platforms. To explore on a broader scale, we selected 12 digital payments from 5 countries of UK, US, Japan, Korea, and Taiwan. With the interplays of user digital behaviors and payment touch points, we group the study cases into four types: (1) Channel Initiated: users originated from retailers with high access to in-store shopping with face-to-face guidance for payment adoption. Providers offer rewards for customer loyalty and secure the retailer’s efficient cash flow management. (2) Social Media Dependent: users usually are digital natives with high access to social media or the internet who shop and pay digitally. Providers might not own physical or online shops but are licensed to aggregate money flows through virtual ecosystems. (3) Early Life Engagement: digital banks race to capture the next generation from popularity to profitability. This type of payment aimed to give children a taste of financial freedom while letting parents track their spending. Providers are to capitalize on the digital payment and e-commerce boom and hold on to new customers into adulthood. (4) Traditional Banking: plastic credit cards are purposely designed as a control group to track the evolvement of business strategies in digital payments. Traditional credit card users may follow the bank’s digital strategy to land on different types of digital wallets or mostly keep using plastic credit cards. This research analyzed business growth models and inter-firms’ coopetition strategies of the selected cases. Results of the multiple case analysis reveal that channel initiated payments bundled rewards with retailer’s business discount for recurring purchases. They also extended other financial services, such as insurance, to fulfill customers’ new demands. Contrastively, social media dependent payments developed new usages and new value creation, such as P2P money transfer through network effects among the virtual social ties, while early life engagements offer virtual banking products to children who are digital natives but overlooked by incumbents. It has disrupted the banking business domains in preparation for the metaverse economy. Lastly, the control group of traditional plastic credit cards has gradually converted to a BaaS (banking as a service) model depending on customers’ preferences. The multi-homing behavior is not avoidable in digital payment competitions. Payment providers may encounter multiple waves of a multi-homing threat after a short period of success. A dynamic cross-business collaboration strategy should be explored to continuously evolve the digital ecosystems and allow users for a broader shopping experience and continual usage.

Keywords: digital payment, digital ecosystems, multihoming users, cross business strategy, user digital behavior intentions

Procedia PDF Downloads 158

Authors: Yu-Lin Chen, Kimberly Koester, Marissa Raymond-Flesh, Anika Thapar, Jay Thapar

Abstract:

Purpose: Effectively communicating adolescent health topics to teens and their parents is crucial. This study emphasizes critically evaluating the optimal use of artificial intelligence tools (AI), which are increasingly prevalent in disseminating health information. By fostering a deeper understanding of AI voice preference in the context of health, the research aspires to have a ripple effect, enhancing the collective health literacy and decision-making capabilities of both teenagers and their parents. This study explores AI voices' potential within health learning modules for annual well-child visits. We aim to identify preferred voice characteristics and understand factors influencing perceived trustworthiness, ultimately aiming to improve health literacy and decision-making in both demographics. Methods: A cross-sectional study assessed preferences and trust perceptions of AI voices in learning modules among teens (11-18) and their parents/guardians in Northern California. The study involved the development of four distinct learning modules covering various adolescent health-related topics, including general communication, sexual and reproductive health communication, parental monitoring, and well-child check-ups. Participants were asked to evaluate eight AI voices across the modules, considering a set of six factors such as intelligibility, naturalness, prosody, social impression, trustworthiness, and overall appeal, using Likert scales ranging from 1 to 10 (the higher, the better). They were also asked to select their preferred choice of voice for each module. Descriptive statistics summarized participant demographics. Chi-square/t-tests explored differences in voice preferences between groups. Regression models identified factors impacting the perceived trustworthiness of the top-selected voice per module. Results: Data from 104 participants (teen=63; adult guardian = 41) were included in the analysis. The mean age is 14.9 for teens (54% male) and 41.9 for the parent/guardian (12% male). At the same time, similar voice quality ratings were observed across groups, and preferences varied by topic. For instance, in general communication, teens leaned towards young female voices, while parents preferred mature female tones. Interestingly, this trend reversed for parental monitoring, with teens favoring mature male voices and parents opting for mature female ones. Both groups, however, converged on mature female voices for sexual and reproductive health topics. Beyond preferences, the study delved into factors influencing perceived trustworthiness. Interestingly, social impression and sound appeal emerged as the most significant contributors across all modules, jointly explaining 71-75% of the variance in trustworthiness ratings. Conclusion: The study emphasizes the importance of catering AI voices to specific audiences and topics. Social impression and sound appeal emerged as critical factors influencing perceived trustworthiness across all modules. These findings highlight the need to tailor AI voices by age and the specific health information being delivered. Ensuring AI voices resonate with both teens and their parents can foster their engagement and trust, ultimately leading to improved health literacy and decision-making for both groups. Limitations and future research: This study lays the groundwork for understanding AI voice preferences for teenagers and their parents in healthcare settings. However, limitations exist. The sample represents a specific geographic location, and cultural variations might influence preferences. Additionally, the modules focused on topics related to well-child visits, and preferences might differ for more sensitive health topics. Future research should explore these limitations and investigate the long-term impact of AI voice on user engagement, health outcomes, and health behaviors.

Keywords: artificial intelligence, trustworthiness, voice, adolescent

Procedia PDF Downloads 55

Authors: Sireetorn Kuharat, Anwar Beg

Abstract:

Motivation- Solar energy constitutes the most promising renewable energy source on earth. Nanofluids are a very successful family of engineered fluids, which contain well-dispersed nanoparticles suspended in a stable base fluid. The presence of metallic nanoparticles (e.g. gold, silver, copper, aluminum etc) significantly improves the thermo-physical properties of the host fluid and generally results in a considerable boost in thermal conductivity, density, and viscosity of nanofluid compared with the original base (host) fluid. This modification in fundamental thermal properties has profound implications in influencing the convective heat transfer process in solar collectors. The potential for improving solar collector direct absorber efficiency is immense and to gain a deeper insight into the impact of different metallic nanoparticles on efficiency and temperature enhancement, in the present work, we describe recent computational fluid dynamics simulations of an annular solar collector system. The present work studies several different metallic nano-particles and compares their performance. Methodologies- A numerical study of convective heat transfer in an annular pipe solar collector system is conducted. The inner tube contains pure water and the annular region contains nanofluid. Three-dimensional steady-state incompressible laminar flow comprising water- (and other) based nanofluid containing a variety of metallic nanoparticles (copper oxide, aluminum oxide, and titanium oxide nanoparticles) is examined. The Tiwari-Das model is deployed for which thermal conductivity, specific heat capacity and viscosity of the nanofluid suspensions is evaluated as a function of solid nano-particle volume fraction. Radiative heat transfer is also incorporated using the ANSYS solar flux and Rosseland radiative models. The ANSYS FLUENT finite volume code (version 18.1) is employed to simulate the thermo-fluid characteristics via the SIMPLE algorithm. Mesh-independence tests are conducted. Validation of the simulations is also performed with a computational Harlow-Welch MAC (Marker and Cell) finite difference method and excellent correlation achieved. The influence of volume fraction on temperature, velocity, pressure contours is computed and visualized. Main findings- The best overall performance is achieved with copper oxide nanoparticles. Thermal enhancement is generally maximized when water is utilized as the base fluid, although in certain cases ethylene glycol also performs very efficiently. Increasing nanoparticle solid volume fraction elevates temperatures although the effects are less prominent in aluminum and titanium oxide nanofluids. Significant improvement in temperature distributions is achieved with copper oxide nanofluid and this is attributed to the superior thermal conductivity of copper compared to other metallic nano-particles studied. Important fluid dynamic characteristics are also visualized including circulation and temperature shoots near the upper region of the annulus. Radiative flux is observed to enhance temperatures significantly via energization of the nanofluid although again the best elevation in performance is attained consistently with copper oxide. Conclusions-The current study generalizes previous investigations by considering multiple metallic nano-particles and furthermore provides a good benchmark against which to calibrate experimental tests on a new solar collector configuration currently being designed at Salford University. Important insights into the thermal conductivity and viscosity with metallic nano-particles is also provided in detail. The analysis is also extendable to other metallic nano-particles including gold and zinc.

Keywords: heat transfer, annular nanofluid solar collector, ANSYS FLUENT, metallic nanoparticles

Procedia PDF Downloads 143

Authors: Nanine Fouche

Abstract:

The Cape Flats is a low-lying sand-covered expanse of approximately 460 square kilometres, situated to the southeast of the central business district of Cape Town in the Western Cape of South Africa. The aeolian sands masking this area are often loose and compressible in the upper 1m to 1.5m of the surface, and there is a general exceedance of the maximum allowable settlement in these sands. The settlement of shallow foundations on Cape Flats sands is commonly predicted using the results of in-situ tests such as the SPT or DPSH due to the difficulty of retrieving undisturbed samples for laboratory testing. Varying degrees of accuracy and reliability are associated with these methods. More recently, shear wave velocity (Vs) profiles obtained from seismic testing, such as continuous surface wave tests (CSW), are being used for settlement prediction. Such predictions have the advantage of considering non-linear stress-strain behaviour of soil and the degradation of stiffness with increasing strain. CSW tests are rarely executed in the Cape Flats, whereas SPT’s are commonly performed. For this reason, and to facilitate better settlement predictions in Cape Flats sand, equations representing shear wave velocity (Vs) as a function of SPT blow count (N60) and vertical effective stress (v’) were generated by statistical regression of site investigation data. To reveal the most appropriate method of overburden correction, analyses were performed with a separate overburden term (Pa/σ’v) as well as using stress corrected shear wave velocity and SPT blow counts (correcting Vs. and N60 to Vs1and (N1)60respectively). Shear wave velocity profiles and SPT blow count data from three sites masked by Cape Flats sands were utilised to generate 80 Vs-SPT N data pairs for analysis. Investigated terrains included sites in the suburbs of Athlone, Muizenburg, and Atlantis, all underlain by windblown deposits comprising fine and medium sand with varying fines contents. Elastic settlement analysis was also undertaken for the Cape Flats sands, using a non-linear stepwise method based on small-strain stiffness estimates, which was obtained from the best Vs-N60 model and compared to settlement estimates using the general elastic solution with stiffness profiles determined using Stroud’s (1989) and Webb’s (1969) SPT N60-E transformation models. Stroud’s method considers strain level indirectly whereasWebb’smethod does not take account of the variation in elastic modulus with strain. The expression of Vs. in terms of N60 and Pa/σv’ derived from the Atlantis data set revealed the best fit with R2 = 0.83 and a standard error of 83.5m/s. Less accurate Vs-SPT N relations associated with the combined data set is presumably the result of inversion routines used in the analysis of the CSW results showcasing significant variation in relative density and stiffness with depth. The regression analyses revealed that the inclusion of a separate overburden term in the regression of Vs and N60, produces improved fits, as opposed to the stress corrected equations in which the R2 of the regression is notably lower. It is the correction of Vs and N60 to Vs1 and (N1)60 with empirical constants ‘n’ and ‘m’ prior to regression, that introduces bias with respect to overburden pressure. When comparing settlement prediction methods, both Stroud’s method (considering strain level indirectly) and the small strain stiffness method predict higher stiffnesses for medium dense and dense profiles than Webb’s method, which takes no account of strain level in the determination of soil stiffness. Webb’s method appears to be suitable for loose sands only. The Versak software appears to underestimate differences in settlement between square and strip footings of similar width. In conclusion, settlement analysis using small-strain stiffness data from the proposed Vs-N60 model for Cape Flats sands provides a way to take account of the non-linear stress-strain behaviour of the sands when calculating settlement.

Keywords: sands, settlement prediction, continuous surface wave test, small-strain stiffness, shear wave velocity, penetration resistance

Procedia PDF Downloads 175

Authors: Jolene M. Helena, Annie M. Joubert, Peace Mabeta, Magdalena Coetzee, Roy Lakier, Anne E. Mercier

Abstract:

Targeting the distant tumour and its microenvironment whilst preserving bone density is important in improving the outcomes of patients with bone metastases. 2-Ethyl-3-O-sulphamoyl-estra1,3,5(10)16-tetraene (ESE-16) is an in-silico-designed 2- methoxyestradiol analogue which aimed at enhancing the parent compound’s cytotoxicity and providing a more favourable pharmacokinetic profile. In this study, the potential radiosensitization effects of ESE-16 were investigated in an in vitro bone metastasis model consisting of murine pre-osteoblastic (MC3T3-E1) and pre-osteoclastic (RAW 264.7) bone cells, metastatic prostate (DU 145) and breast (MDA-MB-231) cancer cells, as well as human umbilical vein endothelial cells (HUVECs). Cytotoxicity studies were conducted on all cell lines via spectrophotometric quantification of 3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide. The experimental set-up consisted of flow cytometric analysis of cell cycle progression and apoptosis detection (Annexin V-fluorescein isothiocyanate) to determine the lowest ESE-16 and radiation doses to induce apoptosis and significantly reduce cell viability. Subsequent experiments entailed a 24-hour low-dose ESE-16-exposure followed by a single dose of radiation. Termination proceeded 2, 24 or 48 hours thereafter. The effect of the combination treatment was investigated on osteoclasts via tartrate-resistant acid phosphatase (TRAP) activity- and actin ring formation assays. Tumour cell experiments included investigation of mitotic indices via haematoxylin and eosin staining; pro-apoptotic signalling via spectrophotometric quantification of caspase 3; deoxyribonucleic acid (DNA) damage via micronuclei analysis and histone H2A.X phosphorylation (γ-H2A.X); and Western blot analyses of bone morphogenetic protein-7 and matrix metalloproteinase-9. HUVEC experiments included flow cytometric quantification of cell cycle progression and free radical production; fluorescent examination of cytoskeletal morphology; invasion and migration studies on an xCELLigence platform; and Western blot analyses of hypoxia-inducible factor 1-alpha and vascular endothelial growth factor receptor 1 and 2. Tumour cells yielded half-maximal growth inhibitory concentration (GI50) values in the nanomolar range. ESE-16 concentrations of 235 nM (DU 145) and 176 nM (MDA-MB-231) and a radiation dose of 4 Gy were found to be significant in cell cycle and apoptosis experiments. Bone and endothelial cells were exposed to the same doses as DU 145 cells. Cytotoxicity studies on bone cells reported that RAW 264.7 cells were more sensitive to the combination treatment than MC3T3-E1 cells. Mature osteoclasts were more sensitive than pre-osteoclasts with respect to TRAP activity. However, actin ring morphology was retained. The mitotic arrest was evident in tumour and endothelial cells in the mitotic index and cell cycle experiments. Increased caspase 3 activity and superoxide production indicated pro-apoptotic signalling in tumour and endothelial cells. Increased micronuclei numbers and γ-H2A.X foci indicated increased DNA damage in tumour cells. Compromised actin and tubulin morphologies and decreased invasion and migration were observed in endothelial cells. Western blot analyses revealed reduced metastatic and angiogenic signalling. ESE-16-induced radiosensitization inhibits metastatic signalling and tumour cell survival whilst preferentially preserving bone cells. This low-dose combination treatment strategy may promote the quality of life of patients with metastatic bone disease. Future studies will include 3-dimensional in-vitro and murine in-vivo models.

Keywords: angiogenesis, apoptosis, bone metastasis, cancer, cell migration, cytoskeleton, DNA damage, ESE-16, radiosensitization.

Procedia PDF Downloads 162

Authors: Atish Bagchi, Siva Chandrasekaran

Abstract:

Background: According to World Bank national accounts data, the estimated global manufacturing value-added output in 2020 was 13.74 trillion USD. These manufacturing processes are monitored, modelled, and controlled by advanced, real-time, computer-based systems, e.g., Industrial IoT, PLC, SCADA, etc. These systems measure and manipulate a set of physical variables, e.g., temperature, pressure, etc. Despite the use of IoT, SCADA etc., in manufacturing, studies suggest that unplanned downtime leads to economic losses of approximately 864 billion USD each year. Therefore, real-time, accurate detection, classification and prediction of machine behaviour are needed to minimise financial losses. Although vast literature exists on time-series data processing using machine learning, the challenges faced by the industries that lead to unplanned downtimes are: The current algorithms do not efficiently handle the high-volume streaming data from industrial IoTsensors and were tested on static and simulated datasets. While the existing algorithms can detect significant 'point' outliers, most do not handle contextual outliers (e.g., values within normal range but happening at an unexpected time of day) or subtle changes in machine behaviour. Machines are revamped periodically as part of planned maintenance programmes, which change the assumptions on which original AI models were created and trained. Aim: This research study aims to deliver a Graph Neural Network(GNN)based hybrid forecasting model that interfaces with the real-time machine control systemand can detect, predict machine behaviour and behavioural changes (anomalies) in real-time. This research will help manufacturing industries and utilities, e.g., water, electricity etc., reduce unplanned downtimes and consequential financial losses. Method: The data stored within a process control system, e.g., Industrial-IoT, Data Historian, is generally sampled during data acquisition from the sensor (source) and whenpersistingin the Data Historian to optimise storage and query performance. The sampling may inadvertently discard values that might contain subtle aspects of behavioural changes in machines. This research proposed a hybrid forecasting and classification model which combines the expressive and extrapolation capability of GNN enhanced with the estimates of entropy and spectral changes in the sampled data and additional temporal contexts to reconstruct the likely temporal trajectory of machine behavioural changes. The proposed real-time model belongs to the Deep Learning category of machine learning and interfaces with the sensors directly or through 'Process Data Historian', SCADA etc., to perform forecasting and classification tasks. Results: The model was interfaced with a Data Historianholding time-series data from 4flow sensors within a water treatment plantfor45 days. The recorded sampling interval for a sensor varied from 10 sec to 30 min. Approximately 65% of the available data was used for training the model, 20% for validation, and the rest for testing. The model identified the anomalies within the water treatment plant and predicted the plant's performance. These results were compared with the data reported by the plant SCADA-Historian system and the official data reported by the plant authorities. The model's accuracy was much higher (20%) than that reported by the SCADA-Historian system and matched the validated results declared by the plant auditors. Conclusions: The research demonstrates that a hybrid GNN based approach enhanced with entropy calculation and spectral information can effectively detect and predict a machine's behavioural changes. The model can interface with a plant's 'process control system' in real-time to perform forecasting and classification tasks to aid the asset management engineers to operate their machines more efficiently and reduce unplanned downtimes. A series of trialsare planned for this model in the future in other manufacturing industries.

Keywords: GNN, Entropy, anomaly detection, industrial time-series, AI, IoT, Industry 4.0, Machine Learning

Procedia PDF Downloads 150

Authors: Stephan Treuke

Abstract:

The paper analyses the neighbourhood effects on the economic mobility of the inhabitants of three segregated communities of Salvador (Brazil), in other words, the socio-economic advantages and disadvantages affecting the lives of poor people due to their embeddedness in specific socio-residential contexts. Recent studies performed in Brazilian metropolis have concentrated on the structural dimensions of negative externalities in order to explain neighbourhood-level variations in a field of different phenomena (delinquency, violence, access to the labour market and education) in spatial isolated and socially homogeneous slum areas (favelas). However, major disagreement remains whether the contiguity between residents of poor neighbourhoods and higher-class condominio-dwellers provides structures of opportunities or whether it fosters socio-spatial stigmatization. Based on a set of interviews, investigating the variability of interpersonal networks and their activation in the struggle for economic inclusion, the study confirms that the proximity of Nordeste de Amaralina to middle-/upper-class communities affects positively the access to labour opportunities. Nevertheless, residential stigmatization, as well as structures of social segmentation, annihilate these potentials. The lack of exposition to individuals and groups extrapolating from the favela’s social, educational and cultural context restricts the structures of opportunities to local level. Therefore, residents´ interpersonal networks reveal a high degree of redundancy and localism, based on bonding ties connecting family and neighbourhood members. The resilience of segregational structures in Plataforma contributes to the naturalization of social distance patters. It’s embeddedness in a socially homogeneous residential area (Subúrbio Ferroviário), growing informally and beyond official urban politics, encourages the construction of isotopic patterns of sociability, sharing the same values, social preferences, perspectives and behaviour models. Whereas it’s spatial isolation correlates with the scarcity of economic opportunities, the social heterogeneity of Fazenda Grande II interviewees and the socialising effects of public institutions mitigate the negative repercussions of segregation. The networks’ composition admits a higher degree of heterophilia and a greater proportion of bridging ties accounting for the access to broader information actives and facilitating economic mobility. The variability observed within the three different scenarios urges to reflect about the responsability of urban politics when it comes to the prevention or consolidation of the social segregation process in Salvador. Instead of promoting the local development of the favela Plataforma, public housing programs priorize technocratic habitational solutions without providing the residents’ socio-economic integration. The impact of negative externalities related to the homogeneously poor neighbourhood is potencialized in peripheral areas, turning its’ inhabitants socially invisible, thus being isolated from other social groups. The example of Nordeste de Amaralina portrays the failing interest of urban politics to bridge the social distances structuring the brazilian society’s rigid stratification model, founded on mecanisms of segmentation (unequal access to labour market and education system, public transport, social security and law protection) and generating permanent conflicts between the two socioeconomically distant groups living in geographic contiguity. Finally, in the case of Fazenda Grande II, the public investments in both housing projects and complementary infrastructure (e.g. schools, hospitals, community center, police stations, recreation areas) contributes to the residents’ socio-economic inclusion.

Keywords: economic mobility, neighborhood effects, Salvador, segregation

Procedia PDF Downloads 279

Authors: Ali Kazemi

Abstract:

Those within the volatile and interconnected international economic markets, appropriately predicting market trends, hold substantial fees for traders and financial establishments. Traditional device mastering strategies have made full-size strides in forecasting marketplace movements; however, monetary data's complicated and networked nature calls for extra sophisticated processes. This observation offers a groundbreaking method for monetary marketplace prediction that leverages the synergistic capability of Graph Convolutional Networks (GCNs) and Long Short-Term Memory (LSTM) networks. Our suggested algorithm is meticulously designed to forecast the traits of inventory market indices and cryptocurrency costs, utilizing a comprehensive dataset spanning from January 1, 2015, to December 31, 2023. This era, marked by sizable volatility and transformation in financial markets, affords a solid basis for schooling and checking out our predictive version. Our algorithm integrates diverse facts to construct a dynamic economic graph that correctly reflects market intricacies. We meticulously collect opening, closing, and high and low costs daily for key inventory marketplace indices (e.g., S&P 500, NASDAQ) and widespread cryptocurrencies (e.g., Bitcoin, Ethereum), ensuring a holistic view of marketplace traits. Daily trading volumes are also incorporated to seize marketplace pastime and liquidity, providing critical insights into the market's shopping for and selling dynamics. Furthermore, recognizing the profound influence of the monetary surroundings on financial markets, we integrate critical macroeconomic signs with hobby fees, inflation rates, GDP increase, and unemployment costs into our model. Our GCN algorithm is adept at learning the relational patterns amongst specific financial devices represented as nodes in a comprehensive market graph. Edges in this graph encapsulate the relationships based totally on co-movement styles and sentiment correlations, enabling our version to grasp the complicated community of influences governing marketplace moves. Complementing this, our LSTM algorithm is trained on sequences of the spatial-temporal illustration discovered through the GCN, enriched with historic fee and extent records. This lets the LSTM seize and expect temporal marketplace developments accurately. Inside the complete assessment of our GCN-LSTM algorithm across the inventory marketplace and cryptocurrency datasets, the version confirmed advanced predictive accuracy and profitability compared to conventional and opportunity machine learning to know benchmarks. Specifically, the model performed a Mean Absolute Error (MAE) of 0.85%, indicating high precision in predicting day-by-day charge movements. The RMSE was recorded at 1.2%, underscoring the model's effectiveness in minimizing tremendous prediction mistakes, which is vital in volatile markets. Furthermore, when assessing the model's predictive performance on directional market movements, it achieved an accuracy rate of 78%, significantly outperforming the benchmark models, averaging an accuracy of 65%. This high degree of accuracy is instrumental for techniques that predict the course of price moves. This study showcases the efficacy of mixing graph-based totally and sequential deep learning knowledge in economic marketplace prediction and highlights the fee of a comprehensive, records-pushed evaluation framework. Our findings promise to revolutionize investment techniques and hazard management practices, offering investors and economic analysts a powerful device to navigate the complexities of cutting-edge economic markets.

Keywords: financial market prediction, graph convolutional networks (GCNs), long short-term memory (LSTM), cryptocurrency forecasting

Procedia PDF Downloads 65