- full-title:: LEV4REC: A feature-based approach to engineering RSSEs type:: [[JournalPaper]] external-links:: [MDEGroup/SoSyM-LEV4REC (github.com)](https://github.com/MDEGroup/SoSyM-LEV4REC) [LEV4REC_Sosym_Reviews - Google Docs](https://docs.google.com/document/d/1y1KsDSZUju1MnMyvZM41YYDhyGvQBzVZzq52_qxRbdU/edit#heading=h.20hmz923tktr) todoist:: https://todoist.com/showTask?id=6718800976 year:: 2023 tentative-submission-deadline:: [[04-09-2023]] status:: [[SUBMITTED]] venue:: [[COLA]] date-submitted:: [[22-11-2023]] priority:: progress:: {{renderer :todomaster}} - Confini del lavoro - Rinominare il tool - #+BEGIN_QUOTE I recommend to add a scenario as a guiding line through the whole article which would explain step [[by]] step how to start]] from exploiting software requirements of the SR to develop and implement them using the low-code approach. #+END_QUOTE - Explanatory / running rec sys developed with and without LEV4REC - We need to clarify the challenges one has to deal with while developing a [[RecSys]] - Probably we are not able to address all of them (report part of the ESEM paper) - The explanatory example should limit the presentation of the challenges that are addressed [[by]] the tool - Put the [[Paper]] in the scope of [[SOSYM]] - We could change the focus (even in the title) on the usage of MDE for supporting the development of [[RecSys]] - Essentially we are showin gan application of MDE for developing tools in the domain of [[RecSys]] - WE claim **domain** independence!!!! - How to convince the reader about such a claim? - The best is to remove / reduce such a clain and list the domains that we can cover. In other words what are the characteristcs of the [[RecSys]] that we support? - We cannot mention specific domains, but we should chracterize the [[RecSys]] (from a component or structure point of view) that we support. For instance, in terms of algorithms we currently support only ..... - On the other end it's important to stress the constraint management aspects (this is the main novelty) - We could even promote this aspect as main goal of the [[Paper]] / approach - The domain concept has to be removed from the [[Paper]]. WE need to show what are the main ingredients of [[RecSys]] and how we enable / manage their composition and configuration - How to answer [[Comments]] like: - ** At this point, I do not know precisely how to transpose the "last lines I already typed before my cursor". Is this enough to use your approach? Or do I need more refined specification? This is where a guiding scenario should be used.** - In my opinion here we should show how the developer is supposed to refine / finalize the [[code]] generated [[by]] the approach. How to accept user customizations / development ? - ** I think that RQ1 is enough for the validation if you are more systematic about "existing recommender systems". ** - systematic means that the considered [[recsys]] are different with respect to some peculiar characteristics that should be introduced when writing the content replacing the domain aspect above - #+BEGIN_QUOTE Il target del lavoro e' uno sviluppatore di sistemi di raccomandazione #+END_QUOTE - Quali sono le challenges che attualmente deve affrontare? - Difficolta nello gestire le varie dipendenze tra algoritmi, dati, encoding, pre-processi (noi la risolviamo con il feature e vincoli) - ### Related works - ![Softw Pract Exp - 2022 - Savary‐Leblanc - Software assistants in software engineering A systematic mapping study.pdf](../assets/Softw_Pract_Exp_-_2022_-_Savary‐Leblanc_-_Software_assistants_in_software_engineering_A_systematic_mapping_study_1684241801009_0.pdf) - ### Tasks - DONE Lavorare su abstract [[01-09-2023]] - DONE RIvedere introduction [[01-09-2023]] - DONE Rivedere sesione 2 [[01-09-2023]] - DONE Rivedere sezione 3 [[01-09-2023]] - DONE Rivedere sezione 4 [[02-09-2023]] - DONE Rivedere sezione 5 [[02-09-2023]] - DONE Rivedere sezione 6 [[02-09-2023]] - DONE Rivedere sezione 7 [[04-09-2023]] - DONE Rivedere sezione 8 [[04-09-2023]] -