Google Summer of Code 2026 proposed ideas - Ιστορικό εκδόσεων

Jimmys στις 15:48, 23 Απριλίου 2026

2026-04-23T15:48:55Z

← Παλαιότερη αναθεώρηση		Αναθεώρηση της 15:48, 23 Απριλίου 2026
Γραμμή 316:		Γραμμή 316:
	====Mentors====		====Mentors====

	~~Iraklis Varlamis, Myrsini Ioannou,~~ Nikos Tsekos, Dimitrios Athanasopoulos		Nikos Tsekos, Dimitrios Athanasopoulos, Myrsini Ioannou, Iraklis Varlamis

	==''' FOSSBot Platform: Simulation Enhancements and AI Integration '''==		==''' FOSSBot Platform: Simulation Enhancements and AI Integration '''==

Jimmys: /* Mentors */

2026-03-10T13:56:00Z

Mentors

← Παλαιότερη αναθεώρηση		Αναθεώρηση της 13:56, 10 Μαρτίου 2026
Γραμμή 285:		Γραμμή 285:
	====Mentors====		====Mentors====

	~~Panagiotis Skarvelis,~~ Dimitrios Athanasopoulos, Nikos Tsekos		Dimitrios Athanasopoulos, Nikos Tsekos, Panagiotis Skarvelis

	=='''GlossAPI: ML-assisted Anonymization Layer and Targeted Pipeline Improvements for Greek Datasets'''==		=='''GlossAPI: ML-assisted Anonymization Layer and Targeted Pipeline Improvements for Greek Datasets'''==

Ellak-editor: /* Mentors */

2026-03-07T17:04:51Z

Mentors

← Παλαιότερη αναθεώρηση		Αναθεώρηση της 17:04, 7 Μαρτίου 2026
Γραμμή 316:		Γραμμή 316:
	====Mentors====		====Mentors====

	Myrsini Ioannou, Nikos Tsekos, Dimitrios Athanasopoulos		Iraklis Varlamis, Myrsini Ioannou, Nikos Tsekos, Dimitrios Athanasopoulos

	==''' FOSSBot Platform: Simulation Enhancements and AI Integration '''==		==''' FOSSBot Platform: Simulation Enhancements and AI Integration '''==

Ellak-editor: /* Mentors */

2026-03-07T17:00:32Z

Mentors

← Παλαιότερη αναθεώρηση		Αναθεώρηση της 17:00, 7 Μαρτίου 2026
Γραμμή 285:		Γραμμή 285:
	====Mentors====		====Mentors====

	Dimitrios Athanasopoulos, Nikos Tsekos		Panagiotis Skarvelis, Dimitrios Athanasopoulos, Nikos Tsekos

	=='''GlossAPI: ML-assisted Anonymization Layer and Targeted Pipeline Improvements for Greek Datasets'''==		=='''GlossAPI: ML-assisted Anonymization Layer and Targeted Pipeline Improvements for Greek Datasets'''==

Dds: /* Related Repositories */ Add GH issue pointer

2026-03-01T14:50:39Z

Related Repositories: Add GH issue pointer

← Παλαιότερη αναθεώρηση		Αναθεώρηση της 14:50, 1 Μαρτίου 2026
Γραμμή 43:		Γραμμή 43:
	====Related Repositories====		====Related Repositories====

	https://github.com/dspinellis/cscout		* https://github.com/dspinellis/cscout
			* Discuss ideas and pose questions at https://github.com/dspinellis/cscout/issues/82

	====Knowledge Prerequisites====		====Knowledge Prerequisites====

Pkst στις 06:56, 19 Φεβρουαρίου 2026

2026-02-19T06:56:28Z

← Παλαιότερη αναθεώρηση		Αναθεώρηση της 06:56, 19 Φεβρουαρίου 2026
Γραμμή 4:		Γραμμή 4:

	For practical information, developers should visit this '''[https://summerofcode.withgoogle.com/how-it-works page]'''.		For practical information, developers should visit this '''[https://summerofcode.withgoogle.com/how-it-works page]'''.


	~~=='''Municipal Budget and Technical Program Visualization Tool'''==~~

	~~====Brief Explanation====~~

	All 332 Greek municipalities publish annual budgets (προϋπολογισμοί) and technical programs (τεχνικά προγράμματα) as PDFs — hundreds of pages of financial data and infrastructure projects that citizens struggle to understand. This project will build a standalone tool that processes these PDFs, extracts structured data, and presents it through an accessible interface. Users will be able to search projects, track spending over time, compare municipalities, and visualize budget allocations. The tool will handle the varied PDF formats used across municipalities and be deployable for all 332 Greek municipalities with minimal configuration.

	This tool addresses a fundamental transparency gap in Greek local government. While budget data is technically public, it is practically inaccessible to citizens due to its format and complexity. OpenCouncil's network of 10+ municipal partners provides immediate access to real budget documents for development and testing, and a clear path to adoption across Greece. The tool will be fully open-source and designed for reuse by other civic tech initiatives. This project complements OpenCouncil's core mission of making municipal governance transparent and accessible.
	~~====Expected Results====~~

	~~1. PDF processing pipeline: Build robust extraction system that handles various PDF formats and structures, validate data quality~~
	~~2.Data model and API: Design database schema, create API endpoints, implement search and comparison features~~
	~~3. User interface and deployment: Build accessible frontend for exploring budgets and projects, deploy for multiple municipalities, write documentation~~

	~~====Duration of the Project====~~

	~~Large Project – 350 hrs~~

	~~====Related Repositories====~~

	~~https://github.com/schemalabz/opencouncil~~

	~~====Knowledge Prerequisites====~~

	~~Python, PDF processing, Data extraction, PostgreSQL, Next.js/React, Data visualization, greek language required~~
	~~====Mentors====~~

	~~Andreas Kouloumos (andreas@opencouncil.gr), Christos Porios (christos@opencouncil.gr)~~



Γραμμή 221:		Γραμμή 191:

	Simon Vellas		Simon Vellas


			=='''Fine-tuning AI transcription models for Greek Municipal Councils'''==

			====Brief Explanation====

			OpenCouncil has accumulated hundreds of hours of Greek municipal council recordings with human-corrected transcriptions and utterance-level timestamps. Greek municipal speech presents unique challenges: regional accents, domain-specific legal and administrative terminology, multiple speakers, and varying audio quality. This project will fine-tune Whisper (or similar open-source speech recognition models) on our dataset to significantly improve transcription accuracy for Greek municipal contexts. The contributor will preprocess training data, experiment with fine-tuning approaches, evaluate model performance against our current pipeline, and deploy the improved model to production. Success means measurably better transcription quality — especially for technical terms and speaker identification.

			OpenCouncil is a civic tech platform that makes Greek municipal council proceedings accessible to citizens through AI-powered transcription, summarization, and search. We have established partnerships with 10+ municipalities across Greece, providing both real-world data and direct validation channels for this project. Our existing dataset of human-corrected transcriptions is a unique resource that makes this fine-tuning project feasible and impactful. The resulting model will be open-sourced and can benefit the entire Greek-speaking NLP community.

			====Expected Results====
			A preprocessed, validated training dataset of Greek municipal speech with utterance-level alignments, ready for model fine-tuning and future research use. A fine-tuned Whisper model that demonstrates measurable improvement in Word Error Rate (WER) over the baseline, particularly on domain-specific terminology (legal, administrative), regional accents, and multi-speaker segments. An evaluation framework with benchmarks and test sets for ongoing quality measurement of Greek municipal transcription. Production integration of the fine-tuned model into OpenCouncil's transcription pipeline, validated on real council meetings from our 10+ municipal partners. Documentation of the fine-tuning methodology, dataset preparation process, and results — contributing to the broader Greek NLP and open-source speech recognition community.

			====Duration of the Project====

			Large Project – 350 hrs

			====Related Repositories====

			https://github.com/schemalabz/opencouncil

			====Knowledge Prerequisites====

			Python, Machine Learning, PyTorch/TensorFlow, Speech Recognition, Greek language required
			====Mentors====

			Andreas Kouloumos (andreas@opencouncil.gr), Christos Porios (christos@opencouncil.gr)





			=='''Municipal Budget and Technical Program Visualization Tool'''==

			====Brief Explanation====

			All 332 Greek municipalities publish annual budgets (προϋπολογισμοί) and technical programs (τεχνικά προγράμματα) as PDFs — hundreds of pages of financial data and infrastructure projects that citizens struggle to understand. This project will build a standalone tool that processes these PDFs, extracts structured data, and presents it through an accessible interface. Users will be able to search projects, track spending over time, compare municipalities, and visualize budget allocations. The tool will handle the varied PDF formats used across municipalities and be deployable for all 332 Greek municipalities with minimal configuration.

			This tool addresses a fundamental transparency gap in Greek local government. While budget data is technically public, it is practically inaccessible to citizens due to its format and complexity. OpenCouncil's network of 10+ municipal partners provides immediate access to real budget documents for development and testing, and a clear path to adoption across Greece. The tool will be fully open-source and designed for reuse by other civic tech initiatives. This project complements OpenCouncil's core mission of making municipal governance transparent and accessible.
			====Expected Results====

			1. PDF processing pipeline: Build robust extraction system that handles various PDF formats and structures, validate data quality
			2.Data model and API: Design database schema, create API endpoints, implement search and comparison features
			3. User interface and deployment: Build accessible frontend for exploring budgets and projects, deploy for multiple municipalities, write documentation

			====Duration of the Project====

			Large Project – 350 hrs

			====Related Repositories====

			https://github.com/schemalabz/opencouncil

			====Knowledge Prerequisites====

			Python, PDF processing, Data extraction, PostgreSQL, Next.js/React, Data visualization, greek language required
			====Mentors====

			Andreas Kouloumos (andreas@opencouncil.gr), Christos Porios (christos@opencouncil.gr)

Pkst στις 06:50, 19 Φεβρουαρίου 2026

2026-02-19T06:50:10Z

← Παλαιότερη αναθεώρηση		Αναθεώρηση της 06:50, 19 Φεβρουαρίου 2026
Γραμμή 4:		Γραμμή 4:

	For practical information, developers should visit this '''[https://summerofcode.withgoogle.com/how-it-works page]'''.		For practical information, developers should visit this '''[https://summerofcode.withgoogle.com/how-it-works page]'''.


			=='''Municipal Budget and Technical Program Visualization Tool'''==

			====Brief Explanation====

			All 332 Greek municipalities publish annual budgets (προϋπολογισμοί) and technical programs (τεχνικά προγράμματα) as PDFs — hundreds of pages of financial data and infrastructure projects that citizens struggle to understand. This project will build a standalone tool that processes these PDFs, extracts structured data, and presents it through an accessible interface. Users will be able to search projects, track spending over time, compare municipalities, and visualize budget allocations. The tool will handle the varied PDF formats used across municipalities and be deployable for all 332 Greek municipalities with minimal configuration.

			This tool addresses a fundamental transparency gap in Greek local government. While budget data is technically public, it is practically inaccessible to citizens due to its format and complexity. OpenCouncil's network of 10+ municipal partners provides immediate access to real budget documents for development and testing, and a clear path to adoption across Greece. The tool will be fully open-source and designed for reuse by other civic tech initiatives. This project complements OpenCouncil's core mission of making municipal governance transparent and accessible.
			====Expected Results====

			1. PDF processing pipeline: Build robust extraction system that handles various PDF formats and structures, validate data quality
			2.Data model and API: Design database schema, create API endpoints, implement search and comparison features
			3. User interface and deployment: Build accessible frontend for exploring budgets and projects, deploy for multiple municipalities, write documentation

			====Duration of the Project====

			Large Project – 350 hrs

			====Related Repositories====

			https://github.com/schemalabz/opencouncil

			====Knowledge Prerequisites====

			Python, PDF processing, Data extraction, PostgreSQL, Next.js/React, Data visualization, greek language required
			====Mentors====

			Andreas Kouloumos (andreas@opencouncil.gr), Christos Porios (christos@opencouncil.gr)

Myrsiniioannou: /* Mentors */

2026-02-17T07:48:55Z

Mentors

← Παλαιότερη αναθεώρηση		Αναθεώρηση της 07:48, 17 Φεβρουαρίου 2026
Γραμμή 225:		Γραμμή 225:
	====Mentors====		====Mentors====

	~~Dimitris~~ Athanasopoulos, Nikos Tsekos		Dimitrios Athanasopoulos, Nikos Tsekos

	=='''GlossAPI: ML-assisted Anonymization Layer and Targeted Pipeline Improvements for Greek Datasets'''==		=='''GlossAPI: ML-assisted Anonymization Layer and Targeted Pipeline Improvements for Greek Datasets'''==
Γραμμή 256:		Γραμμή 256:
	====Mentors====		====Mentors====

	Myrsini Ioannou, Nikos Tsekos, ~~Dimitris~~ Athanasopoulos		Myrsini Ioannou, Nikos Tsekos, Dimitrios Athanasopoulos

	==''' FOSSBot Platform: Simulation Enhancements and AI Integration '''==		==''' FOSSBot Platform: Simulation Enhancements and AI Integration '''==

Pkst: /* Python bindings for Apothesis= */

2026-02-12T08:46:12Z

Python bindings for Apothesis=

← Παλαιότερη αναθεώρηση		Αναθεώρηση της 08:46, 12 Φεβρουαρίου 2026
Γραμμή 443:		Γραμμή 443:
	Diomidis Spinellis		Diomidis Spinellis

	='''Python bindings for Apothesis'''==		='''Python bindings for Apothesis'''=

	====Brief Explanation====		====Brief Explanation====

Pkst στις 08:44, 12 Φεβρουαρίου 2026

2026-02-12T08:44:20Z

← Παλαιότερη αναθεώρηση		Αναθεώρηση της 08:44, 12 Φεβρουαρίου 2026
Γραμμή 443:		Γραμμή 443:
	Diomidis Spinellis		Diomidis Spinellis

			='''Python bindings for Apothesis'''==

			====Brief Explanation====

			This project focuses on creating modern Python bindings for Apothesis to seamlessly connect its high-performance C++ core with Python-based AI workflows. While the existing C++ implementation delivers speed and efficiency for large-scale computation, the lack of native Python integration limits accessibility for researchers and machine learning practitioners who primarily work within Python ecosystems such as NumPy, PyTorch, and TensorFlow. The bindings will expose Apothesis’ core functionality through a clean, Pythonic API, preserving performance while ensuring memory safety and ease of installation. By bridging efficient C++ algorithms with Python’s flexibility and extensive AI tooling, this work enables rapid experimentation, smoother integration into machine learning pipelines, and broader adoption within the research community.

			====Expected Results====
			The expected results of the project are the delivery of fully functional Python bindings for Apothesis that provide seamless access to its C++ core from Python. The bindings will expose the main computational components through a clean and intuitive API, while preserving the performance and efficiency of the underlying C++ implementation.The project will also include comprehensive documentation, usage examples, and unit tests to ensure reliability, maintainability, and ease of adoption. Overall, the outcome will be a stable, extensible interface that lowers the barrier to entry for researchers and developers, enabling faster experimentation, improved interoperability with machine learning pipelines, and broader community engagement.

			====Duration of the Project====

			Large Project – 350 hrs

			====Related Repositories====

			https://github.com/nixeimar/Apothesis

			====Knowledge Prerequisites====
			C++, python

			====Mentors====

			Cheimarios Nikolaos, Vissarion Fysikopoulos

	[[Κατηγορία:GSOC]]		[[Κατηγορία:GSOC]]

			[[Category:GSOC]]