Merge pull request #99 from garandria/gh-pages-edit

fix typo of last talks

content/talks/20240613-diverse-coffee.md

@@ -3,11 +3,32 @@ date: 2024-06-13T13:00:00
 title: "Large language models as oracles for instantiating ontologies with domain-specific knowledge"
 abstract: >
 
- Endowing intelligent systems with semantic data commonly requires designing and instantiating ontologies with domain-specific knowledge. Especially in the early phases, those activities are typically performed manually by human experts possibly leveraging on their own experience. The resulting process is therefore time-consuming, error-prone, and often biased by the personal background of the ontology designer. Objective. To mitigate that issue, we propose a novel domain-independent approach to automatically instantiate ontologies with domain-specific knowledge, by leveraging on large language models (LLMs) as oracles. Method. Starting from (i) an initial schema composed by inter-related classes andproperties and (ii) a set of query templates, our method queries the LLM multiple times, and generates instances for both classes and properties from its replies. Thus, the ontology is automatically filled with domain-specific knowledge, compliant to the initial schema. As a result, the ontology is quickly and automatically enriched with manifold instances, which experts may consider to keep, adjust, discard, or complement according to their own needs and expertise. Contribution. We formalise our method in general way and instantiate it over various LLMs, as well as on a concrete case study. We report experiments rooted in the nutritional domain where an ontology of food meals and their ingredients is semi-automatically instantiated from scratch, starting from a categorisation of meals and their relationships. There, we analyse the quality of the generated ontologies and compare ontologies attained by exploiting different LLMs. Finally, we provide a SWOT analysis of the proposed method.
-
-https://gciatto.github.io/presentation-llm4kg/#/
+  Background. Endowing intelligent systems with semantic data commonly requires
+  designing and instantiating ontologies with domain-specific
+  knowledge. Especially in the early phases, those activities are typically
+  performed manually by human experts possibly leveraging on their own
+  experience. The resulting process is therefore time-consuming, error-prone, and
+  often biased by the personal background of the ontology designer. Objective. To
+  mitigate that issue, we propose a novel domain-independent approach to
+  automatically instantiate ontologies with domain-specific knowledge, by
+  leveraging on large language models (LLMs) as oracles. Method. Starting from (i)
+  an initial schema composed by inter-related classes andproperties and (ii) a set
+  of query templates, our method queries the LLM multiple times, and generates
+  instances for both classes and properties from its replies. Thus, the ontology
+  is automatically filled with domain-specific knowledge, compliant to the initial
+  schema. As a result, the ontology is quickly and automatically enriched with
+  manifold instances, which experts may consider to keep, adjust, discard, or
+  complement according to their own needs and expertise. Contribution. We
+  formalise our method in general way and instantiate it over various LLMs, as
+  well as on a concrete case study. We report experiments rooted in the
+  nutritional domain where an ontology of food meals and their ingredients is
+  semi-automatically instantiated from scratch, starting from a categorisation of
+  meals and their relationships. There, we analyse the quality of the generated
+  ontologies and compare ontologies attained by exploiting different
+  LLMs. Finally, we provide a SWOT analysis of the proposed method.
 
 event: DiverSE Coffee
 location: Rennes, France
-speaker: Giovanni Ciatto
----
+speaker: Giovanni Ciatto (University of Bologna)
+url_slides: https://gciatto.github.io/presentation-llm4kg/#/
+---

content/talks/20240627-diverse-coffee.md

@@ -2,23 +2,44 @@
 date: 2024-06-27T13:00:00
 title: "Internships"
 abstract: >
-
   ### Development of a Web SCADA interface for the control of a Fischertechnik-based factory, by Thierry Traore
- 
- In modern industry, user interfaces play a crucial role in the management of complex supervisory control and data acquisition (SCADA) systems. This project focuses on the design of a web interface using Angular and Spring Boot for a Fischertechnik-based plant, aiming to optimize the management and control of industrial processes through its digital twin.
 
-We will explore the technical challenges encountered during the development of the interface, including the integration of advanced functionalities such as near real-time visualization of production data, remote control of equipment and alert management. We will also present the strategies adopted to ensure the architecture's scalability.
+  In modern industry, user interfaces play a crucial role in the management of
+  complex supervisory control and data acquisition (SCADA) systems. This project
+  focuses on the design of a web interface using Angular and Spring Boot for a
+  Fischertechnik-based plant, aiming to optimize the management and control of
+  industrial processes through its digital twin.
 
-In conclusion, this project demonstrates the effectiveness of using Angular and Spring Boot together to create a modern, reactive and high-performance user interface for a Fischertechnik plant. It illustrates how these technologies can significantly improve the productivity and reliability of industrial operations through an optimized interface.
 
+  We will explore the technical challenges encountered during the development of
+  the interface, including the integration of advanced functionalities such as
+  near real-time visualization of production data, remote control of equipment and
+  alert management. We will also present the strategies adopted to ensure the
+  architecture's scalability.
 
-  ### Optimizing Weather and Water Data Management: Integration, Automated Cleaning, and Dashboard Visualization, by Aubry Tonnerre
 
+  In conclusion, this project demonstrates the effectiveness of using Angular and
+  Spring Boot together to create a modern, reactive and high-performance user
+  interface for a Fischertechnik plant. It illustrates how these technologies can
+  significantly improve the productivity and reliability of industrial operations
+  through an optimized interface.
+
+
+  ### Optimizing Weather and Water Data Management: Integration, Automated Cleaning, and Dashboard Visualization, by Aubry Tonnerre
 
-  During my internship at IRISA, I will present my academic background and describe the work I am undertaking this summer as part of the DiverSE team.
+  During my internship at IRISA, I will present my academic background and
+  describe the work I am undertaking this summer as part of the DiverSE team.
 
-Due to developments in tree production in Laos, the country's geography has undergone significant changes. In response to these changes, a research team established a watershed in the Houay Pano region of Laos 20 years ago. This watershed is utilized to measure water flow based on rainfall and the water level. However, the data collected is occasionally erroneous, complicating its usability. Consequently, human intervention is required to correct these discrepancies. The objective of my project is to design a system to automate the data cleaning process and develop an IoT architecture using LORA technologies to enhance the existing system.
 
+  Due to developments in tree production in Laos, the country's geography has
+  undergone significant changes. In response to these changes, a research team
+  established a watershed in the Houay Pano region of Laos 20 years ago. This
+  watershed is utilized to measure water flow based on rainfall and the water
+  level. However, the data collected is occasionally erroneous, complicating its
+  usability. Consequently, human intervention is required to correct these
+  discrepancies. The objective of my project is to design a system to automate the
+  data cleaning process and develop an IoT architecture using LORA technologies to
+  enhance the existing system.
 
 event: DiverSE Coffee
 location: Rennes, France