Novel Time Slicing Approach for Customer Defection Models
In this session Kyriakos Georgiou, Chief Risk and Data Science Officer at Finloup, will delve into the topic of forecasting customer churn in a non-contractual setting. Taking inspiration from the latest developments in machine learning that incorporate multiple time intervals, we present an inventive approach for training and evaluating known as the sliding multi-time slicing (SMTS) technique.
Kyriakos Georgiou
Kyriakos Georgiou
Chief Risk and Data Science Officer
Date
Thursday, November 2, 2023
13:00 -
14:00 EEST
Track
Workshop
Share
Facebook
LinkedIn

Speaker

Kyriakos Georgiou
Chief Risk and Data Science Officer

Connect with Kyriakos Georgiou on:
Kyriakos Georgiou

From the Same Track

November, 2
16:30 -
16:50 EEST
Workshop
How A.I. Makes the Network More Secure, Smarter and Simpler
The attack surface is expanding as the IT environment undergoes significant changes. Cloud adoption is increasing, remote and hybrid users now dominate, and IoT devices are becoming more prevalent on corporate networks.

Many organizations rely on a complex array of products that were not designed to support today's highly distributed environment. Additionally, as breach vectors multiply, enterprise application stacks become more complex, and new threats emerge daily, the time and resources required for businesses to remain secure and operational have grown exponentially.

In this session, Evi Kechagia, Cybersecurity Technical Solutions Architect at Cisco Greece, Cyprus and Malta will examine how teams can proactively identify threats before they become incidents within their organizations. This can be achieved by harnessing the power of Artificial Intelligence and Machine Learning models through pattern recognition. We will also explore how teams can establish secure and resilient networks and streamline operations by utilizing tools and systems that simplify the tasks of IT and Security teams. These techniques are particularly crucial in situations where data analysis is complex and resource-intensive.
Evi Kechagia
Evi Kechagia
Cybersecurity Technical Solutions Architect, Cisco Greece, Cyprus and Malta
Business Intelligence
Intermediate

Share this Session

November, 2
15:50 -
16:50 EEST
Workshop
Quantifying the Incremental Impact of an Agri-tech Product with Causal AI
To effectively market a digital product, it's crucial to understand its impact on key user metrics and determine its level of success. In this session, George Giannarakis Data Scientist at National Observatory of Athens and Ilias Tsoumas Research Data Scientist at National Observatory of Athens will discuss the importance of this causal question, the challenges in providing rigorous answers, and introduce causal AI as a powerful tool to address these challenges.

The session will illustrate these concepts through the story of a digital agriculture product developed by the National Observatory of Athens in collaboration with Corteva Agriscience for cotton farmers in Greece. The comprehensive case study covers all aspects of the project, from product development to on-field deployment. We'll specifically focus on how causal AI techniques were used to accurately estimate the incremental impact of the product on the farmers' yield, which served as the primary measure of success.
Ilias Tsoumas
Ilias Tsoumas
Research Data Scientist, National Observatory of Athens
George Giannakaris
Georgios Giannarakis
Data Scientist, National Observatory of Athens
A.I. in Product Management
Basic

Share this Session

November, 2
13:30 -
13:50 EEST
Quick Session
Unleashing the Potential of AI: Exploring Business Opportunities and Challenges
In this session, Elisavet Kyritsi, the Alliances Director EMEA at Teradate, will explore the ever-changing realm of Artificial Intelligence (AI) and its significant influence on fostering business innovation.

Elisavet will delve into the intricacies of harnessing the potential of generative AI, shedding light on its role and limitations. Additionally, we will demonstrate how to unlock the full potential of data analytics and machine learning in conjunction with generative AI to propel corporate performance and bolster competitiveness. Lastly, we will dispel prevalent misconceptions and tackle the obstacles that arise when implementing generative AI in practical scenarios.
Elisavet Kyritsi
Elisavet Kyritsi
Alliances Director EMEA, Teradata
Business Intelligence
Basic

Share this Session

November, 2
10:25 -
11:25 EEST
Workshop
Leveraging Large Language Models in a Communication Platform
In the ever-changing world of communication platforms, the integration of AI has completely transformed our interactions and collaborations. In this session, we will explore how AI can empower teams to streamline processes and increase engineering productivity in various domains. We will specifically focus on areas such as Site Reliability Engineering, Incident Management, Change Management, and more.

One of the main concerns surrounding AI integration is data privacy. In this session, Spiros Economakis Head of Cloud Infrastructure at Mattermost, will discuss how organizations can harness the power of AI without compromising the security of their private data or sharing it with third parties. Additionally, the session will delve into the role of Open Source AI Models.

In an era defined by collaboration and collective innovation, open-source models have become essential tools for making AI accessible to all. We will uncover how these models can be utilized within a communication platform, allowing for customization and adaptation to meet specific organizational needs.
Spiros Economakis
Spiros Economakis
Head of Cloud Infrastructure, Mattermost
Generative A.I.
Intermediate

Share this Session

Scroll to Top