| Ambos lados, revisión anteriorRevisión previaPróxima revisión | Revisión previa |
| temas_previos [2022/03/30 22:37] – jbarbay | temas_previos [2022/04/01 23:38] (actual) – ahogan |
|---|
| El semestre indica el semestre del 'F' (del trabajo de memoria en sí). | El semestre indica el semestre del 'F' (del trabajo de memoria en sí). |
| |
| ===== Otoño 2022 ====== | ===== Primavera 2022 ====== |
| |
| | * **Objetivo**: //Autocompletando preguntas sobre Wikidata// |
| | * **Profesor Guía**: [[ahogan@dcc.uchile.cl|Aidan Hogan]] |
| | * **Descripción**: [[https://www.wikidata.org/wiki/Wikidata:Main_Page|Wikidata]] es una gran fuente de información que permite contestar muchos tipos de preguntas. Pero para hacer consultar complejas sobre Wikidata, hay que ser experto en SPARQL, algo que limita su alcance. El objetivo de este trabajo es desarrollar un sistema de "question answering" (QA) sobre Wikidata. La tarea QA asume una pregunta en lenguaje natural como por ejemplo "¿Cuándo fue estrenada la película Sharknado?", y debería dar una respuesta directa como "2013". La tarea general es compleja, pero este tema de memoria tiene que ver con desarrollar una versión más simple de QA basada en autocompletar preguntas sobre un catálogo existente de preguntas y consultas. Por ejemplo, al teclar "cuando fue", se generan sugerencias de preguntas que empiezan así, y si se selecciona "¿Cuándo fue estrenada la película [PELÍCULA]?", se permite también buscar una película para reemplazar "[PELÍCULA]" usando autocomplete. Basado en las selecciones del usuario, se genera una consulta SPARQL que se evalúa sobre el servicio de consulta de Wikidata, generando los resultados finales. [Todavía disponible: 2022-03-22] |
| |
| * **Objetive**: //Perma-Simulation//: An HTML5 simulation of Andrew Millison's sandbox for teaching Permaculture online {{ :populous.png?200|}} | * **Objetive**: //Perma-Simulation//: An HTML5 simulation of Andrew Millison's sandbox for teaching Permaculture online {{ :populous.png?200|}} |
| * **Code Available**: https://github.com/msakuta/WebPopulous [2014], running on http://msakuta.github.io/WebPopulous/populous.html | * **Code Available**: https://github.com/msakuta/WebPopulous [2014], running on http://msakuta.github.io/WebPopulous/populous.html |
| * **Description**: (//Memoria//) In his videos for the Permaculture course at Oregon State University (e.g. https://www.youtube.com/watch?v=V3tpaIf6Jcc), Andrew Millison uses a sandbox combined with a webcam and data projector to illustrate visually various interactions between water, land and cultures, showing the passing of years in a matter of minutes. Currently, the next step for his students is to move to the real world and slow time, where mistakes are discovered years later. Could current web technology, such as demonstrated in the remake of the Populous game http://msakuta.github.io/WebPopulous/populous.html allowing to raise and lower lands while independant agents react to the changing state of the terrain, be used to procure students a cheap (i.e. running into any modern browser) interactive (more than watching a video) experience in a virtual world where they can put in practice Andrew Millison's lessons in accelerated time? We propose to develop a simplistic simulator video game in Javascript and HTML5 which can be parameterized to complement each video of Andrew Millison's course with a practical exercise where the student can put into practice the concepts described in the video, in their own virtual sandbox. The success of the project will be evaluated in terms of usability and extendibility 1) by Andrew Millison himself on three exercises, based on three selected videos from his course, 2) by a set of volunteer students from his course on such exercises, and 3) by a web developper from Oregon State University charged to configure the simulator to create additional exercises for other videos of the course, if time allows. The student must be comfortable having meetings and writing their report in English, as the project will be in collaboration with Andrew Millison. [Last Revision: [2022-03-28 Mon]] | * **Description**: (//Memoria//) In his videos for the Permaculture course at Oregon State University (e.g. https://www.youtube.com/watch?v=V3tpaIf6Jcc), Andrew Millison uses a sandbox combined with a webcam and data projector to illustrate visually various interactions between water, land and cultures, showing the passing of years in a matter of minutes. Currently, the next step for his students is to move to the real world and slow time, where mistakes are discovered years later. Could current web technology, such as demonstrated in the remake of the Populous game http://msakuta.github.io/WebPopulous/populous.html allowing to raise and lower lands while independant agents react to the changing state of the terrain, be used to procure students a cheap (i.e. running into any modern browser) interactive (more than watching a video) experience in a virtual world where they can put in practice Andrew Millison's lessons in accelerated time? We propose to develop a simplistic simulator video game in Javascript and HTML5 which can be parameterized to complement each video of Andrew Millison's course with a practical exercise where the student can put into practice the concepts described in the video, in their own virtual sandbox. The success of the project will be evaluated in terms of usability and extendibility 1) by Andrew Millison himself on three exercises, based on three selected videos from his course, 2) by a set of volunteer students from his course on such exercises, and 3) by a web developper from Oregon State University charged to configure the simulator to create additional exercises for other videos of the course, if time allows. The student must be comfortable having meetings and writing their report in English, as the project will be in collaboration with Andrew Millison. [Last Revision: [2022-03-28 Mon]] |
| |
| |
| |
| ===== Primavera 2022 ====== | |
| |
| |
| |
| |
| * **Objetivo**: Generación de software web para visualización interactiva de mallas geométricas Lepp. | * **Objetivo**: Generación de software web para visualización interactiva de mallas geométricas Lepp. |
