BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Resource Scheduling and Allocation in HPC Cyberinfrastructure
DTSTART;VALUE=DATE-TIME:20241204T092000Z
DTEND;VALUE=DATE-TIME:20241204T094000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2493@events.chpc.ac.za
DESCRIPTION:Speakers: Olalekan Samuel  Ogunleye* (University of Mpumalanga
 )\nAbstract\nHigh-Performance Computing (HPC) systems play a pivotal role 
 in modern scientific research\, enabling complex simulations\, data analys
 is\, and large-scale modelling across disciplines such as climate science\
 , genomics\, physics\, and engineering. As these systems grow in scale and
  sophistication\, the efficient scheduling and allocation of computational
  resources become crucial for ensuring optimal system performance\, maximi
 sing resource utilisation\, and meeting the needs of diverse user communit
 ies. In HPC environments\, resource scheduling and allocation determine ho
 w tasks are assigned to hardware resources such as CPUs\, GPUs\, memory\, 
 storage\, and network bandwidth. Effective scheduling strategies are criti
 cal for maintaining fairness among users\, optimising job throughput\, red
 ucing waiting times\, and enhancing energy efficiency.\nCyberinfrastructur
 e\, the integration of advanced computing platforms with large-scale data 
 storage and high-speed networks\, addlayers of complexity to resource mana
 gement. The heterogeneity of hardware\, dynamic workload demands\, and mul
 ti-user environments require advanced resource scheduling algorithms. Trad
 itional approaches like First-Come\, First-Served (FCFS)\, Shortest Job Fi
 rst (SJF)\, and Backfilling have evolved to meet these challenges\, while 
 more advanced strategies like Priority Scheduling\, Gang Scheduling\, and 
 Hybrid Scheduling offer increased flexibility and efficiency. Energy-aware
  scheduling has also gained importance\, given the significant portion of 
 operational costs that energy consumption in large HPC systems can account
  for.\n\nhttps://events.chpc.ac.za/event/139/contributions/2493/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2493/
END:VEVENT
BEGIN:VEVENT
SUMMARY:SAWS Climate Services and Infrastructure
DTSTART;VALUE=DATE-TIME:20241204T090000Z
DTEND;VALUE=DATE-TIME:20241204T092000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2423@events.chpc.ac.za
DESCRIPTION:Speakers: Dawn Mahlobo* (South African Weather Service)\nThe S
 outh African Weather Service (SAWS) is a key part of the country's weather
  system. SAWS runs a complex system of tools to observe weather. This incl
 udes manual weather stations\, automatic rain measuring stations\, automat
 ic weather stations\, weather radars\, a network that measures sunlight\, 
 and a system to detect lightning. These systems give important information
  right away for predicting the weather and studying the climate. SAWS uses
  advanced computer models to create detailed weather and ocean forecasts. 
 These models are important for predicting extreme weather events like cycl
 ones and coastal storms\, which can seriously affect communities and busin
 esses. The SAWS climate infrastructure is a changing system that supports 
 the country's work to fight against and adjust to climate change. South Af
 rica is improving its ability to watch for\, predict\, and react to climat
 e-related problems by using new technologies\, strong research efforts\, a
 nd working with internal and international communities.\n\nhttps://events.
 chpc.ac.za/event/139/contributions/2423/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2423/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Empowering Data-Driven Research with AI Tools
DTSTART;VALUE=DATE-TIME:20241204T135000Z
DTEND;VALUE=DATE-TIME:20241204T141000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2488@events.chpc.ac.za
DESCRIPTION:Speakers: Lungile Nkosi* (University of Pretoria)\nBackground\
 nThe growing volume and complexity of data\, particularly in health and so
 cial research\, present significant challenges\, particularly in terms of 
 data security and access to secure datasets. These issues are compounded w
 hen working with vulnerable populations\, exposing data to potential cyber
 security risks. AI-powered tools like Chisquares are addressing these chal
 lenges by embedding advanced security features—such as encryption\, data
  protection compliance\, and secure storage—ensuring data safety while p
 romoting inclusivity and accessibility for researchers in high-risk enviro
 nments.\n\nMethods\nThis study examines how AI-powered tools\, with Chisqu
 ares as a case study\, enhance data security and streamline research workf
 lows. It highlights advanced security measures\, including encryption\, co
 mpliance with global data protection regulations\, and flexible data stora
 ge options. We also demonstrate how these tools reduce the technical burde
 n on researchers while maintaining data integrity and privacy.\n\nResults\
 nAI-powered platforms like Chisquares enhance research workflows by provid
 ing secure data access and storage through features such as encryption\, r
 ole-based access controls\, and data protection compliance. These platform
 s also support offline functionality\, ensuring secure data handling in ar
 eas with limited internet access. Furthermore\, over 80% of tasks\, from d
 ata cleaning to manuscript preparation\, are automated\, with built-in saf
 eguards for accuracy and security. By integrating data management into a s
 ingle platform\, these tools mitigate the risks associated with transferri
 ng data across multiple systems. \n\nConclusion\nAI-powered tools like Chi
 squares are transforming research by embedding security throughout the pro
 cess. They provide secure data access\, offline functionality\, and flexib
 le data storage options\, empowering researchers to handle sensitive infor
 mation with confidence. These tools not only protect data but also enable 
 impactful\, data-driven research\, particularly in resource-constrained an
 d high-risk settings\, ultimately supporting better public health outcomes
 .\n\nhttps://events.chpc.ac.za/event/139/contributions/2488/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2488/
END:VEVENT
BEGIN:VEVENT
SUMMARY:AcousNomaly: Learning to Detect Anomalies in Acoustic Telemetry Da
 ta using Machine learning and Deep Learning
DTSTART;VALUE=DATE-TIME:20241204T133000Z
DTEND;VALUE=DATE-TIME:20241204T135000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2419@events.chpc.ac.za
DESCRIPTION:Speakers: Siphendulwe Zaza* (MSc Applied Mathematics student f
 rom Rhodes University)\nAcoustic telemetry data plays a vital role in unde
 rstanding the be-\nhaviour and movement of aquatic animals. However\, thes
 e datasets\,\nwhich can often consist of millions of individual data point
 s\, often\ncontain anomalous detections that can pose challenges in data a
 nalysis\nand interpretation. Anomalies in acoustic telemetry data can occu
 r due\nto various biological and environmental factors\, and technological
  limi-\ntations. Anomalous movements are generally identified manually\, w
 hich\ncan be extremely time-consuming in large datasets. As such\, this st
 udy\nfocuses on automating the process of anomaly detection in telemetry\n
 datasets using machine learning (ML) and artificial intelligence (AI)\nmod
 els. Fifty dusky kob *(Argyrosomus japonicus)* were surgically fit-\nted w
 ith unique coded acoustic transmitters in the Breede Estuary\,\nSouth Afri
 ca\, and their movements were monitored using an array of\n16 acoustic rec
 eivers deployed throughout the estuary between 2016\nand 2021\, resulting 
 in more than 3 million individual data points. The\nresearch approach comb
 ined the use of Neural Network (NN) models\nand autoencoders to construct 
 an efficient anomaly detection system. The model is proficient at learning
  the normal movement patterns within\nthe data\, effectively distinguishin
 g between normal and anomalous be-\nhaviour\, and exceeding 90% across all
  four evaluation metrics including\naccuracy\, precision\, recall\, and F1
 . However\, it may encounter chal-\nlenges in accurately detecting anomali
 es where they deviate slowly from\nthe expected movement patterns. Despite
  this limitation\, the model\ndemonstrates promising capabilities by pinpo
 inting the precise loca-\ntions of anomalous entries within the dataset. F
 urther investigation\,\nincluding refinement and optimization of the model
 ’s parameters and\ntraining process\, especially with memory-based NN-AE
 \, may enhance\nits ability to detect anomalies with greater accuracy and 
 reliability.\n\nhttps://events.chpc.ac.za/event/139/contributions/2419/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2419/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Data sharing\, fabric and effective governance in the error of big
  data analytics
DTSTART;VALUE=DATE-TIME:20241204T113000Z
DTEND;VALUE=DATE-TIME:20241204T115000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2487@events.chpc.ac.za
DESCRIPTION:Speakers: Mawande Mongo* (Altron)\nIn the era of big data\, or
 ganizations are increasingly leveraging advanced data management strategie
 s to enhance operational efficiency and drive innovation. This paper explo
 res three critical components of modern data management: data sharing\, da
 ta fabric\, and data governance.\nData sharing facilitates seamless acces
 s and exchange of data across different departments and external partners\
 , fostering collaboration and informed decision-making. However\, it also 
 raises concerns about data security and privacy\, necessitating robust mec
 hanisms to ensure data integrity and compliance.\nData fabric represents 
 an architectural approach that integrates various data sources\, both stru
 ctured and unstructured\, into a unified\, intelligent data management fra
 mework. This approach enhances data accessibility\, quality\, and real-tim
 e analytics\, enabling organizations to derive actionable insights from th
 eir data assets.\nData governance encompasses the policies\, procedures\,
  and standards that ensure data is managed effectively and responsibly. It
  addresses issues related to data quality\, privacy\, and compliance\, ens
 uring that data is accurate\, secure\, and used ethically. Effective data 
 governance is crucial for maintaining trust and accountability in data-dri
 ven environments.\nTogether\, these elements form the backbone of a resili
 ent data strategy\, empowering organizations to harness the full potential
  of their data while mitigating risks associated with data misuse and regu
 latory non-compliance\n\nhttps://events.chpc.ac.za/event/139/contributions
 /2487/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2487/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Q&A
DTSTART;VALUE=DATE-TIME:20241204T102000Z
DTEND;VALUE=DATE-TIME:20241204T103000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2494@events.chpc.ac.za
DESCRIPTION:https://events.chpc.ac.za/event/139/contributions/2494/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2494/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Q&A
DTSTART;VALUE=DATE-TIME:20241204T125000Z
DTEND;VALUE=DATE-TIME:20241204T130000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2489@events.chpc.ac.za
DESCRIPTION:https://events.chpc.ac.za/event/139/contributions/2489/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2489/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Transformer-based Sense Embeddings with Deep Learning Large Langua
 ge Models for Low-Resource Language Disambiguation
DTSTART;VALUE=DATE-TIME:20241204T143000Z
DTEND;VALUE=DATE-TIME:20241204T145000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2416@events.chpc.ac.za
DESCRIPTION:Speakers: Hlaudi Masethe* (Tshwane University of Technology)\n
 Determining a word's accurate meaning in each context is known as Word Sen
 se Disambiguation (WSD)[1]\, and it is one of the most significant problem
 s in Natural Language Processing (NLP)[2]. This undertaking is particularl
 y challenging for low-resource languages like Sesotho sa Leboa since there
  are few annotated corpora and linguistic resources available for them. Th
 is study explores the application of many transformer-based and deep learn
 ing models for WSD in Sesotho sa Leboa\, with good results despite the lan
 guage's resource constraints. This study employs a variety of deep learnin
 g architectures\, including transformer-based models such as Recurrent Neu
 ral Networks with Long Short-Term Memory (RNN-LSTM)\, Bidirectional Gated 
 Recurrent Units (BiGRU)\, and an LSTM-based Language Model (LSTMLM)\, as w
 ell as models like DistilBERT with Naive Bayes (DistilBERT & NB)\, DeBERTa
 \, T5\, and ALBERT[3][4]. \n\nThe study makes use of the unique hardware c
 haracteristics of the T4 GPU to improve and optimize the runtime of deep l
 earning language models\, especially big transformers. The purpose of the 
 NVIDIA T4 Tensor Core GPU is to speed up deep learning and machine learnin
 g operations. It works especially well for training and inferring big lang
 uage models. Every phase entail making efficient use of software optimizat
 ions in addition to comprehending and utilizing hardware features. The BiG
 RU model outperformed other deep learning language models with an accuracy
  of 79%\, demonstrating the effectiveness of bidirectional processing effe
 ctively capturing contextual information. With an accuracy of 70%\, DeBERT
 a beat the other transformer-based large language models to enhance pre-tr
 aining techniques that prioritize spatial and contextual embeddings.\n\n**
 Keywords:** Word Sense Disambiguation\, Sesotho sa Leboa\, Low-Resourced L
 anguages\, Deep Learning\, Transformer Models\, RNN-LSTM\, BiGRU\, DeBERTa
 \, NLP\n\nhttps://events.chpc.ac.za/event/139/contributions/2416/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2416/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Obtaining and Using Data for Decision Support and Innovation in Re
 source-Constrained Environments
DTSTART;VALUE=DATE-TIME:20241204T141000Z
DTEND;VALUE=DATE-TIME:20241204T143000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2417@events.chpc.ac.za
DESCRIPTION:Speakers: Audrey Masizana* (University of Botswana)\nWe live i
 n an increasingly data-driven world\, and as academics\, researchers and p
 rofessionals\, we need the ability to manage\, analyze\, and interpret dat
 a efficiently.  We also need to know how to source data and get around the
  various red-tape systems that are even more prevalent in resource-limited
  contexts\, such as in Africa.  This presentation reflects on over a decad
 e of research and practical experience in harnessing data for decision-mak
 ing in education\, healthcare\, and public administration\, particularly w
 ithin the context of Botswana. \n\nOne key area of focus has been optimizi
 ng data management processes to enhance operational efficiency and accurac
 y. My work on Data Matching has tackled a number of practical problems\, i
 ncluding a project which showcases the potential of intelligent data match
 ing techniques to streamline administrative tasks\, prevent errors\, and u
 ltimately improve the quality of educational outcomes\, by matching studen
 t registration records with exam scripts. \n\nIn healthcare\, data-driven 
 decision support systems have proven invaluable. My work on expert systems
  for HIV and AIDS Information and development of Decision Support for Prov
 ision of HIV Treatments has demonstrated the role of expert systems and ar
 tificial intelligence in improving treatment outcomes for complex diseases
  such as HIV/AIDS. This was extended further with a project which analyzed
  drug resistance in HIV/AIDS patients using clustering\, which leverages A
 I to address drug resistance challenges in public health. \n\nThe integrat
 ion of traditional knowledge systems with modern data technologies is anot
 her emerging area of interest. The study on Patient Management and Health 
 Outcome Monitoring by Traditional Healers in Botswana provides a unique pe
 rspective on how data from non-conventional healthcare systems can be inco
 rporated into formal health monitoring frameworks. \n\nThe COVID-19 pandem
 ic underscored the importance of adaptable and scalable data management sy
 stems. In my publication on Experiences\, Lessons\, and Challenges With Ad
 apting REDCap for COVID-19 Laboratory Data Management in a Resource-Limite
 d Country\, we explored the adoption of a research data platform to manage
  critical health data\, highlighting the adaptability of digital tools in 
 crisis situations. \n\nBuilding on this\, I am currently leading a project
  that focuses on predicting COVID-19 mortality in Botswana using machine l
 earning models. By leveraging high-dimensional clinical datasets\, which i
 nclude both structured clinical parameters and unstructured textual data\,
  our models aim to provide early and accurate predictions of patient outco
 mes. This project illustrates the power of machine learning in enhancing h
 ealthcare systems' ability to proactively manage pandemic-related challeng
 es. The insights gained from this work not only inform clinical interventi
 ons but also help optimize resource allocation in healthcare\, a critical 
 need in resource-constrained environments like Botswana. \n\n*Through this
  presentation\, I will synthesize these experiences to explore how data ca
 n drive innovation\, improve decision-making\, and overcome challenges in 
 resource-constrained environments. The talk will also offer insights into 
 future opportunities for data utilization in sectors ranging from educatio
 n to healthcare\, with a focus on low-resource settings\, as well as exper
 iences on hurdles to obtaining data for research and recommendations for b
 etter policies related to data management\, sharing\, and protection.*\n\n
 https://events.chpc.ac.za/event/139/contributions/2417/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2417/
END:VEVENT
BEGIN:VEVENT
SUMMARY:A Secure Data-Centric Model for SMMEs Across Africa: Bringing Stak
 eholders Together
DTSTART;VALUE=DATE-TIME:20241204T123000Z
DTEND;VALUE=DATE-TIME:20241204T125000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2420@events.chpc.ac.za
DESCRIPTION:Speakers: Sonwabo Mdwaba* (Pan African Information Communicati
 on Technology Association)\, Nobert Jere (University of Fort Hare)\nCybers
 ecurity continues to be a threat to many sectors and individuals within Af
 rica. As a result\, Small\, Medium\, and Micro Enterprises (SMMEs) are als
 o affected. SMMEs face numerous challenges related to data security\, part
 icularly as they increasingly rely on digital tools and platforms for thei
 r operations. The situation is worse for SMMEs\, particularly those in rur
 al or underserved areas. There have been differently initiatives currently
  in palace to alert the African community on cybersecurity. However\, impl
 ementing a Secure Data-Centric Model for SMMEs in Africa comes with severa
 l challenges due to the unique socio-economic and technological landscape 
 of the continent. Additionally\, researchers seem to have noticed that cyb
 ersecurity initiatives are mainly dominated by those in the Information Te
 chnology space\, which seems not to be enough. To overcome these challenge
 s\, African SMMEs may need support from governments\, industry bodies\, an
 d international organizations in the form of funding\, training\, and acce
 ss to affordable. Collaborative efforts to raise awareness\, improve regul
 atory frameworks\, and build local cybersecurity capacity are also crucial
  in helping SMMEs in Africa implement effective secure data-centric models
 . Despite growing accessibility\, the cost of advanced cybersecurity solut
 ions can be prohibitive for many SMMEs. \nA systematic review on current c
 ybersecurity data and interviews with cybersecurity experts was done. This
  was supported by the recently completed Pan African Information Communica
 tion Technology Association Cybersecurity Conference held in August 2024. 
 The data from the presenters and may points raised during the conference w
 ere noted and considered for this talk. The main research question for thi
 s talk is:\nHow can secure data-centric model for SMMEs across Africa be d
 esigned through stakeholders engagements? \nResults are clear on the speci
 fic stakeholders that are required\, and the key components of the data ce
 ntric model components are well documented. As an example\, collaboration 
 among these stakeholders—government\, industry\, academia\, and civil so
 ciety—is essential to create an enabling environment where secure data-c
 entric practices can flourish among SMMEs across the continent.  The Secur
 e Data-Centric Model for SMMEs is designed to address these challenges by 
 creating a robust framework that prioritizes data security while fostering
  collaboration among key stakeholders. We argue that the Secure Data-Centr
 ic Model for SMMEs is a comprehensive approach to ensuring data security w
 hile fostering collaboration among key stakeholders. By focusing on secure
  data practices\, stakeholder engagement\, and regulatory compliance\, the
  model empowers SMMEs to thrive in the digital economy while protecting th
 eir most asset i.e. data. In summary\, a Secure Data-Centric Model for SMM
 Es is about creating a comprehensive\, scalable\, and collaborative approa
 ch to data security\, ensuring that small businesses can protect their val
 uable information.\n\nhttps://events.chpc.ac.za/event/139/contributions/24
 20/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2420/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Crafting Open Data for Open Science: Technical Innovation and Data
  Management in Environmental Research — The SAEON Open Data Platform
DTSTART;VALUE=DATE-TIME:20241204T121000Z
DTEND;VALUE=DATE-TIME:20241204T123000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2414@events.chpc.ac.za
DESCRIPTION:Speakers: Mark Jacobson* (South African Environmental Observat
 ion Network)\nCrafting Open Data for Open Science: Technical Innovation an
 d Data\nManagement in Environmental Research — The SAEON Open Data Platf
 orm\n\nThe South African Environmental Observation Network (SAEON) is one 
 of the National\nResearch Foundation (NRF)’s Research Infrastructure Pla
 tforms and serves as a sustained\, coordinated\, responsive and comprehens
 ive South African earth observation network. SAEON delivers long-term\, re
 liable data for scientific research and informs decision-making to support
  a knowledge society and improve quality of life.\n\nThe Open Data Platfor
 m (ODP) is one of SAEON’s research data infrastructure comprising an agg
 regation of databases\, services and web applications that facilitate the 
 preservation\, publication\, discovery\, and dissemination of earth observ
 ation and environmental data in South Africa. The ODP was certified as a t
 rustworthy data repository by CoreTrustSeal in 2023. In the evolving lands
 cape of environmental data\, which continuously changes in response to tec
 hnological\, societal\, and scientific developments\, effective curation a
 nd publication processes are crucial for ensuring dataset accessibility an
 d usability. SAEON’s commitment to the FAIR principles — making data F
 indable\, Accessible\, Interoperable\, and Reusable — guides data manage
 ment and dissemination practices. SAEON ensures that metadata is comprehen
 sive and adheres to established standards\, enabling users to understand t
 he context and quality of the data.\n\nWhile the assemblage of systems con
 stituting the ODP has grown organically over many years\, an information a
 rchitecture has simultaneously evolved to enable the centralised metadata 
 management system to accept data submissions from a variety of sources. It
  permits quality control and value-added annotations by SAEON’s data cur
 ation team. Additionally\, it allows for interoperability with and publica
 tion of metadata to a variety of data cataloguing systems both locally and
  globally. The development of this abstract information architecture has l
 ed to the emergence of useful high-level patterns including many-to-many c
 onnectivity between data producers (archives) and data consumers (catalogu
 es)\, differentiated access control supportive of multitenancy\, and an ex
 tensible ontological framework.\n\nIn this presentation\, we will cover:\n
 \n - SAEON’s approach to managing environmental data publication\, inclu
 ding rigorous curation process\, quality checks\, and the assignment of DO
 Is\;\n - The role of comprehensive metadata in enhancing discoverability a
 nd user engagement\;\n - Characteristics of the ODP information architectu
 re and its utility in real-world use cases.\n\nhttps://events.chpc.ac.za/e
 vent/139/contributions/2414/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2414/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Applications of 4IR on the diagnosis and management of STIs among 
 key populations in Sub-Saharan Africa: A Systematic Review
DTSTART;VALUE=DATE-TIME:20241204T115000Z
DTEND;VALUE=DATE-TIME:20241204T121000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2415@events.chpc.ac.za
DESCRIPTION:Speakers: Claris Siyamayambo* (1South Africa Medical Research 
 Council/ University of Johannesburg (SAMRC/UJ) Pan African Centre for Epid
 emics Research (PACER) Extramural Unit\, Faculty of Health Sciences\, Sout
 h Africa)\nIntroduction: The Fourth Industrial Revolution (4IR) is trendin
 g because of the major transformations it has brought to human life. Artif
 icial intelligence including machine learning are 4IR technologies that ca
 n generate intelligent machines that can be used for the diagnosis and man
 agement of HIV and associated sexually transmitted infections. Key populat
 ions are disproportionately affected by HIV and STIs due to specific risk 
 behaviors\, marginalization\, and structural factors that contribute to a 
 lack of access to health services. The 4IR technologies are used in report
 ing the key populations’ STI vulnerability\, transmission\, and treatmen
 t. \nAim: To explore the use of 4IR technologies in the diagnosis and mana
 gement of STIs for key populations in Sub-Saharan Africa.\nMethods: A revi
 ew of the literature published from 2015 onwards was done. Manual and elec
 tronic searches on various databases including PubMed Central\, SCOPUS\, a
 nd Science Direct were conducted. The Preferred Reporting Items for System
 atic Reviews and meta-analysis statements for protocol guidelines were fol
 lowed\, and the review is registered in the International Prospective Regi
 ster of Systematic Reviews database (Ludwig-Walz\, Dannheim\, Pfadenhauer\
 , Fegert & Bujard\, 2023). PROSPERO Registration ID is CRD42023468734.\nRe
 sults: Different machine learning algorithms including random forest class
 ifier\, support vector machine\, and logistic regression can be used to ge
 nerate models to predict STIs. The 4IR technologies can help to track peop
 le who have accessed STI services including those who have the potential t
 o transmit infections including prevention and care for the sake of enhanc
 ing patient outcomes.\nConclusion: Machine learning models can help identi
 fy individuals at high risk of contracting HIV and assist policymakers in 
 developing targeted HIV prevention and screening strategies informed by so
 cio-demographic and risk behavioural data. There remains a gap in HIV diag
 nosis for key populations. The 4IR technologies can use available data for
  building models on HIV diagnosis and care among key populations in Sub-Sa
 haran Africa and significantly improve elements required to facilitate dia
 gnostic and management approaches.\n\nhttps://events.chpc.ac.za/event/139/
 contributions/2415/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2415/
END:VEVENT
BEGIN:VEVENT
SUMMARY:DIRISA: A National Infrastructure for Research Data Management and
  Collaboration
DTSTART;VALUE=DATE-TIME:20241204T100000Z
DTEND;VALUE=DATE-TIME:20241204T102000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2421@events.chpc.ac.za
DESCRIPTION:Speakers: Katlego W. Phoshoko* (Data Intensive Research Initia
 tive of South Africa (DIRISA)\, CSIR)\nAs part of the global drive for cyb
 erinfrastructure providers to continue enabling\, enhancing and empowering
  research data management through collaborative frameworks\, the Data Inte
 nsive Research Initiative of South Africa (DIRISA)\, as a national researc
 h data initiative\, provides an integrated suite of free tools and service
 s designed to optimize research data workflows\, ultimately amplifying the
  impact of research and scholarly endeavours.\n\nThis presentation will sh
 owcase DIRISA's offerings and tools\, which include cloud storage\, ensuri
 ng that researchers can safely store and manage their data\, research data
  management planning\, seamless data transfers and ways to enhance the dis
 coverability and citation of datasets\, fostering greater visibility in th
 e academic and research landscape.\n\nIn the presentation attendees will e
 xplore how DIRISA can serve as a vital partner in their research journey\,
  paving the way for innovative outcomes and collaborative success\, which 
 also accelerates research impact across multiple disciplines.\n\nhttps://e
 vents.chpc.ac.za/event/139/contributions/2421/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2421/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Data driven decision-making and policy for application in water an
 d sanitation systems
DTSTART;VALUE=DATE-TIME:20241204T094000Z
DTEND;VALUE=DATE-TIME:20241204T100000Z
DTSTAMP;VALUE=DATE-TIME:20260807T152007Z
UID:indico-contribution-657-2418@events.chpc.ac.za
DESCRIPTION:Speakers: Ridhwaan Suliman* (CSIR)\nThis project looks at the 
 use of data to develop evidence-based decision-making and policy for appli
 cation in water and sanitation systems. It includes a review and understan
 ding of local and global databases and resources\, developing data mining 
 and management standards and practices\, and applying data analysis techni
 ques to advance water and sanitation systems in South Africa.\n\nData mini
 ng and management plays an important role in advancing water and sanitatio
 n systems\, ensuring sufficient monitoring and the sustainable delivery of
  essential services. The application of data mining and management encompa
 sses the collection\, processing\, and storage of vast datasets derived fr
 om various sources such as local and global databases\, dashboards\, senso
 r networks\, satellite imagery\, and public health records. \n\nData analy
 sis and modelling techniques then facilitate the identification of pattern
 s\, trends\, and anomalies\, which are important for informed decision-mak
 ing\, strategic planning and public policy. By leveraging predictive analy
 tics and forecasting\, government departments and water management authori
 ties can anticipate demand fluctuations\, optimise resource allocation\, a
 nd enhance the efficiency of water distribution networks. Similarly\, in w
 ater sanitation\, data mining and analysis assists in monitoring system pe
 rformance\, detecting potential failures\, and mitigating health risks by 
 providing early warnings of contamination events. \n\nThe adoption of robu
 st data management frameworks ensures the integration\, storage\, and acce
 ssibility of diverse datasets\, supporting real-time monitoring and long-t
 erm strategic initiatives. Challenges such as data privacy\, accuracy\, an
 d the need for interdisciplinary collaboration need to be addressed to ens
 ure the reliability and efficacy of these systems. The convergence of data
  mining and management in water and sanitation sectors holds significant p
 romise for enhancing operational efficiency\, ensuring resource sustainabi
 lity\, and safeguarding public health. \n\nData integration poses a signif
 icant hurdle due to varying formats and structures across different source
 s. The primary challenge is data quality and accuracy including issues lik
 e missing values and statistical outliers. Water databases may contain a d
 iverse range of data\, including spatial\, temporal\, and multi-dimensiona
 l information. Integrating and reconciling these different types of data c
 an be challenging\, especially when they come from various sources with di
 fferent formats and structures.\n\nhttps://events.chpc.ac.za/event/139/con
 tributions/2418/
LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma
URL:https://events.chpc.ac.za/event/139/contributions/2418/
END:VEVENT
END:VCALENDAR