- - ##  Notes - ## 🤝Meetings - **CROSSMINER GROUP MEETING** collapsed:: true - type:: [[meeting]] people:: [[people/crossminer]] date:: [[18-01-2024]] - 17:41 duration:: - **Work Proposal 1** collapsed:: true - ![image.png](../assets/image_1705595863887_0.png) - ![image.png](../assets/image_1705595895693_0.png) - ![image.png](../assets/image_1705595922644_0.png) - ![image.png](../assets/image_1705596057438_0.png) - This is very much related to [[PROJECTS/MOSAICO]] [[Ideas]] - ![image.png](../assets/image_1705596363893_0.png) - [[Ideas]] - ![image.png](../assets/image_1705596384502_0.png) - ![image.png](../assets/image_1705596428643_0.png) - ![image.png](../assets/image_1705596460118_0.png) - ![image.png](../assets/image_1705596603840_0.png) - Results obtained with cross-validation. - Using readme file to predict only one tag - The dataset is like this, only one tag is assigned to each labeled model - ![image.png](../assets/image_1705596719019_0.png) - We are missing the GitHub link. - - **Work Proposal 2 - Automate scientific workflows** - ![image.png](../assets/image_1705597065372_0.png) - ![image.png](../assets/image_1705597083314_0.png) - ![image.png](../assets/image_1705597099581_0.png) - [[Ideas]] - ![image.png](../assets/image_1705597119998_0.png) - ![image.png](../assets/image_1705597135965_0.png) - ![image.png](../assets/image_1705597483010_0.png) - We need to characterize the source code. The idea was to consider Jupyter notebook in a speculative way instead of Orange models that are limited in size. - We need to characterize Jupyper notebook to do so. WE decided to rely on an existing work that classify Jupyter notebooks. So we could focus on those that implement data pipelines / workflows so that we can train a recommender for data pipelines / workflows. - ![image.png](../assets/image_1705597765627_0.png) - ![image.png](../assets/image_1705597776917_0.png) - ![image.png](../assets/image_1705597823885_0.png){:height 312, :width 1360} - - **MOSAICO meeting** type:: [[meeting]] external-links:: tags:: [[PROJECTS/MOSAICO]] people:: date:: [[18-01-2024]] - 9:00 duration:: 100 collapsed:: true - ### Agenda - **Introduction partners** - DONE @ all: could you please all prepare a presentation of 5 minutes describing your institution and your expertises. background-color:: green - **Concept presentation**: 5 minutes. --> Massimo presentation of the concept. - **Key specific objectif vs SOTA vs WP introduction and challenges** for each of the 4 building blocks composing the MOSAICO platform: - @YORK: may I ask you to present KSO 1 + SOTA related to communication protocol between AI agents in SE tasks + WP1 and possible tasks division --> 10 minutes - DONE @UDA: may I ask you to present KSO 2 + SOTA related to management of agents communities for SE tasks + WP2 and possible tasks division --> 10 minutes background-color:: green - @IMT Massimo: may I ask you to present KSO 3 + SOTA related to coordination and collaboration of agents in SE tasks + WP3 and possible tasks division --> 10 minutes - @LIST Jordi: may I ask you to present KSO 4 + SOTA related to governance in SE tasks + WP4 and possible tasks division --> 10 minutes - [[BESSER]] collapsed:: true - ![image.png](../assets/image_1705567329362_0.png){:height 411, :width 709} - ![image.png](../assets/image_1705567366089_0.png){:height 383, :width 712} - **Contribution of INTRASOFT and ECLIPSE into the project** - @John and ECL partners, please prepare suggestions. - **Discussion about use cases** - Immersion: feedback - *UDA/ Davide*, any feedback on your side ? - INTRASOFT, any end users ? - **Thesis of Moamin** collapsed:: true - type:: [[meeting]] external-links:: tags:: [[SERVICES/PHDICT]] people:: date:: [[18-01-2024]] - 14:54 duration:: - - Data quality - - ![image.png](../assets/image_1705585686036_0.png) - ![image.png](../assets/image_1705585722937_0.png) - How DataNode overlap with deployment components - WHat are the aspects of novelty of this metamodel - ![image.png](../assets/image_1705585812393_0.png) - What's the need for this compared with workflow languages of process based languages like BPMN? - Data pipeline language [[question]] - What's the novelty with respect to - Apache NIFI - Orange - ML workflow technologies - ![image.png](../assets/image_1705585934268_0.png) - ![image.png](../assets/image_1705585996342_0.png) - ![image.png](../assets/image_1705586040213_0.png) - ![image.png](../assets/image_1705586108677_0.png) - What's the novelty of this? - Data Quality management - ![image.png](../assets/image_1705586172229_0.png) - ![image.png](../assets/image_1705586225440_0.png) - ![image.png](../assets/image_1705586260051_0.png) - ![image.png](../assets/image_1705586294317_0.png) - Data quality #question - Is it possible to add new data dimensions - How have you elicited the quality dimensions? - How to define quality, which can be based on a very subjective definitions - ![image.png](../assets/image_1705586349951_0.png) - ![image.png](../assets/image_1705586389788_0.png) - It is not clear what are the research questions you wanted to answer with this evaluation [[question]] - ![image.png](../assets/image_1705586447972_0.png) - ![image.png](../assets/image_1705586491995_0.png) - To appreciate the results it is necessary to understand the research questions - ![image.png](../assets/image_1705586522960_0.png) - ![image.png](../assets/image_1705586540223_0.png) - ![image.png](../assets/image_1705586563568_0.png) - What kind of data you can support - Is you tool able to generate tool or it's just a graphical interface - Why are you considering GPS in you data model - What are your future plans? - WHat are the things that you did not manage to cover during your study and that you think you want to keep working - If you would to teach, would you like to have a dedicated course on this or part of an existing course like software architecture - - - {{query (and <%current page%> (property :type "meeting") (not <%current page%>))}} - ## 📜Reviews - {{query (and [[REVIEWS]] (or (property :date-end <%today%>) (property :date-start <%today%>) (property :date-submitted <%today%>)))}} query-table:: true query-properties:: [:file :year :venue :deadline] - ## ⤴️Omnivore readings - {{query (and (property :source [[Omnivore]]) (or (property :deadline <%today%>) (property :date-saved <%today%>) (property :date-archived <%today%>)))}} query-table:: true query-properties:: [:labels :state :full-title]