Team Data Exploit

<p>The team began by generating around 50 ideas from broad prompts and stakeholder discussions. Projects with clear datasets and defined goals advanced more smoothly, highlighting the importance of structured data. Standardizing data for AI and LLMs was identified as essential for future adaptability.</p>
<p>Developing AI literacy, accessing beginner-friendly tools like Lovable, ChatGPT, Claude, Cursor, and Anything LLM, and starting with simple projects are crucial steps for building momentum. Emphasis was placed on creativity, curiosity, and problem-solving rather than coding skills, with ongoing reflection and refinement using LLMs to guide best practices.</p>
<h4>Key takeaways:</h4>
<ul>
<li>Clear goals and datasets lead to smoother progress.</li>
<li>Data preparation is vital for AI project flexibility.</li>
<li>AI literacy and accessible tools empower all users.</li>
<li>Start small and iterate for project success.</li>
<li>Creativity and curiosity are as important as technical skills.</li>
</ul>

<p>Team Data Exploit developed eight prototypes focused on data-driven tools, including social media text adaptation, text-to-speech podcasting, text-to-image and image-to-video generation, a Chalmers social media map, a research database chatbot, an interactive student journey, and a news evaluator.</p>
<p>The video introduces these projects and invites viewers to explore detailed demo videos for each. Additionally, viewers are encouraged to watch a team discussion about collaborative and exploratory work in data exploitation and AI at Chalmers.</p>
<h4>Key takeaways:</h4>
<ul>
<li>Eight innovative prototypes were created around data and AI.</li>
<li>Tools support social media, accessibility, visualization, and research.</li>
<li>More detailed demo videos and a group discussion are available to watch.</li>
</ul>

<p>The video presents Team Data Exploit’s work on the Formatting Machine, highlighting their development of text-to-speech and podcast generator modules. These tools allow users to convert text, files, or links into spoken audio or podcasts in various languages and tones, enhancing accessibility and educational use. The podcast generator uses ChatGPT to produce scripts and synthetic voices for conversations.</p>
<p>Challenges included integrating advanced voice models and striking a balance between building custom solutions and using existing services. The team envisions broader applications, such as integrating text-to-speech with other media generation tools and creating interactive, voice-driven data platforms. The technology is rapidly advancing, promising more natural voices and functionality soon.</p>
<h4>Key takeaways:</h4>
<ul>
<li>Users can transform text into realistic speech or podcasts in multiple languages.</li>
<li>The podcast generator enables custom conversation styles and roles.</li>
<li>Integration with future technologies could create richer, interactive experiences.</li>
<li>The technology greatly improves accessibility and educational opportunities.</li>
<li>Rapid advancement means features and voice quality will continue improving.</li>
</ul>

<p>The video presents Team Data Exploit’s work on the "Formatting Machine" project, focusing on AI-driven tools for generating educational and marketing materials. Key modules demonstrated include a text-to-image application (using services like DALL-E and LightLLM) that transforms prompts into images, and an image-to-video tool (via Runway) that animates images based on user input. Both tools offer features for customizing style, size, quality, and AI-enhanced prompts, though current API limitations—cost, video length, and input restrictions—are noted.</p>
<p>The presenter discusses the rapid development of generative AI, challenges in scalability, and opportunities for integrating these modules into educational content (e.g., with avatars and slides). The technology is seen as highly promising for communications and teaching, though best practices and impactful uses are still evolving.</p>
<h4>Key takeaways:</h4>
<ul>
<li>AI tools can convert text to images and images to short videos.</li>
<li>Features include prompt enhancement and customization options.</li>
<li>Current limitations involve cost and API constraints.</li>
<li>Future use cases include education, marketing, and internal communication.</li>
<li>The technology is rapidly evolving with significant potential.</li>
</ul>

<p>The video introduces Team Data Exploit’s Chalmers Social Media Map, a prototype tool that aggregates and visualizes social media data about Chalmers University from Meltwater and Reddit. The tool offers interactive exploration via filters, sentiment analysis, word clouds, and geolocation, making it useful for monitoring public perception, recruitment efforts, and identifying discussion trends.</p>
<p>The system can be expanded with more data sources, better sentiment analysis, and timeline visualizations but currently faces challenges with data relevance, sentiment accuracy, and live data integration. It highlights the superiority of custom LLM-based sentiment analysis over commercial solutions and the importance of tailored data exploration for institutional needs.</p>
<h4>Key takeaways:</h4>
<ul>
<li>Aggregates and analyzes social media data about Chalmers University</li>
<li>Includes sentiment analysis and word clouds for insights</li>
<li>Prototype aids communication, recruitment, and reputation management</li>
<li>Custom LLM analysis outperforms commercial sentiment tools</li>
<li>Future improvements include live data and richer audience metrics</li>
</ul>

<p>The video presents Team Data Exploit’s “Formatting Machine” project, designed to help professionals effortlessly reformat texts for various social media platforms and audiences. The tool uses AI to generate tailored versions of a message—adhering to a defined tone such as Chalmers’—for platforms like TikTok, Instagram, and LinkedIn, with potential for translations, accessibility, and educational uses.</p>
<p>The system automates repetitive communication tasks, saving time while allowing for human input at the final stage. Future enhancements may include text simplification, more media outputs, and integration with other tools for review or translation.</p>
<h4>Key takeaways:</h4>
<ul>
<li>The Formatting Machine streamlines multi-platform text adaptation.</li>
<li> AI outputs are customized for tone, audience, and channel.</li>
<li>The tool saves time on repetitive communication tasks.</li>
<li>Translation, simplification, and multimedia features are possible future upgrades.</li>
</ul>

<p>The video presents a gamified tool developed by Team Data Exploit for prospective students to simulate a personalized journey at Chalmers University. The tool uses a large language model (LLM) to generate unique, interactive scenarios informed by fact files about programs, activities, and opportunities at Chalmers. Users make choices at key stages—such as selecting a program, joining activities, and picking a master’s track—resulting in a custom summary of their virtual Chalmers experience.</p>
<p>The project highlights challenges in aggregating accurate data and discusses possible expansions, like career pathways and mentorship matching. It serves as an engaging inspiration tool, with potential to aid both recruitment and student exploration.</p>
<h4>Key takeaways:</h4>
<ul>
<li>Gamified simulation tool lets students explore Chalmers journeys interactively.</li>
<li>LLM creates unique, scenario-based experiences informed by real data.</li>
<li>Could expand to career guidance and alumni connections.</li>
<li>Addresses recruitment and student lifestyle queries effectively.</li>
</ul>

<p>The video showcases Team Data Exploit’s work developing a chatbot interface for Chalmers University’s research database. Using large language models (LLMs) and a “React Agent” capable of multi-step reasoning, the chatbot interacts with structured data on research projects, people, and publications, helping users efficiently find relevant information.</p>
<p>The team discusses challenges such as integrating multiple data sources and building a toolbox that enables the agent to make effective searches across different indices. They explain the potential to enhance research discovery by connecting social graphs, using ORCID profiles, adding network visualizations, and upgrading to modern databases that support advanced integrations. The project serves as a proof of concept for scalable, interactive research data tools.</p>
<h4>Key takeaways:</h4>
<ul>
<li>Chatbot uses LLMs for dynamic database querying and summaries.</li>
<li>Integrates Chalmers research data but can expand with more sources.</li>
<li>Toolbox design and data integration are key challenges.</li>
<li>Social graph and network visualizations enhance discovery.</li>
<li>Future upgrades could boost capability and usability.</li>
</ul>

<p>The video introduces "News Evaluator," an app designed to streamline the process of summarizing and assessing research articles. Users can upload Excel files or enter DOI numbers, after which the tool generates article summaries, potential headlines, and evaluates newsworthiness.</p>
<p>The team discusses challenges such as obtaining full-text access, dealing with PDF formats, and ensuring the quality of AI-generated content. The app aims to automate parts of news evaluation and enhance research accessibility for users, especially at Chalmers.</p>
<h4>Key takeaways:</h4>
<ul>
<li>News Evaluator automates research article assessment.</li>
<li>Users can upload data or input DOIs for analysis.</li>
<li>The tool provides summaries, headlines, and news scores.</li>
<li>Addressing data access and AI quality remains crucial.</li>
<li>The app boosts research accessibility and news evaluation.</li>
</ul>

Team Smart Campus

Team Smart Uni

Team 10xIndividual

Team Education

Team Upskill

Team AI-based feedback

Explorative AI projects 2025

Choose subject

Course syllabus for

Subject

Each academic year contains different courses

See programme page at universityadmissions.se

Educational area

Permissible substitutions of compulsory courses

To see your programme syllabus – select the academic year in which you started the programme.

[en: page.studyprogrammepage.syllabusesview.coursesdescription]

Explorative AI projects 2025

Team Data Exploit

Introduction to Team Data Exploit

Video description: Introduction to Team Data Exploit

Project overview

Video description: Project overview

Text-to-speech

Video description: Text-to-speech

Text-to-image-to-video

Video description: Text-to-image-to-video

Chalmers Life Journey

Video description: Chalmers Life Journey

Research chatbot

Video description: Research chatbot

The News Evaluator

Video description: The News Evaluator

Team Data Exploit

Introduction to Team Data Exploit

Video description: Introduction to Team Data Exploit

Project overview

Video description: Project overview

Text-to-speech

Video description: Text-to-speech

Text-to-image-to-video

Video description: Text-to-image-to-video

Chalmers Social Media Map

Video description: Chalmers Social Media Map

Social Media Text Transformer

Video description: Social Media Text Transformer

Chalmers Life Journey

Video description: Chalmers Life Journey

Research chatbot

Video description: Research chatbot

The News Evaluator

Video description: The News Evaluator