Dartmouth Institute for Health Policy and Clinical Practice
Dartmouth Digital Commons (Dartmouth College)Not a member yet
8214 research outputs found
Sort by
Memenify: Leveraging Machine Learning and Large Language Models for Contextual Meme Creation
No full textMemes are powerful in reflecting cultural trends and humor that resonate with specific audiences. Can artificial intelligence be used to generate memes that people enjoy? Will people appreciate AI’s creative ability when it comes to abstract forms of com- munication like memes? This thesis presents Memenify, a novel system for generating contextually appropriate memes by leveraging Artificial Intelligence. By combining the interpretative strengths of random forests with the text processing capabilities of GPT-4, Memenify produces meme dialogues and monologues tailored to a given context. Through the analysis of meme patterns with the Random Forests classifier, Memenify suggests and predicts appropriate meme templates, guiding the LLM to generate relevant and humorous content to construct a complete meme. To not only validate and test our system, but also evaluate people’s experience with AI powered meme generators, we implemented a website featuring an ’iMessage’-style chat inter- face, powered by a server running our algorithms. Our user study results indicate that Memenify effectively captures users’ context and produces relevant memes that users find relatable. This study demonstrates the potential of AI in meme generation while highlighting key challenges that current AI systems cannot solve, such as generating humor
HeartDJ - Music Recommendation and Generation through Biofeedback from Heart Rate Variability
Full text linkThis study investigates the integration of real-time physiological data with AI-generated music to enhance emotional well-being, stress regulation, and focus, using Heart Rate Variability (HRV) as a biomarker of autonomic function. Conducted in two phases—Stable Audio Open (SAO) and Suno (SUNO)—the research evaluates biofeedback-driven music interventions across varying daily music-listening habits.
In the SAO phase, short AI-generated instrumental tracks were compared with Spotify recommendations and guided meditation. Modest HRV improvements were observed in biofeedback conditions, but participants noted emotional limitations, citing short track lengths and abrupt transitions.
The SUNO phase addressed these limitations with longer, more complex AI-generated compositions combined with mixed Spotify and AI recommendations. Notably, the condition with four AI-generated tracks (0S4G) resulted in a significant HRV decrease (), indicating heightened sympathetic activation and reduced parasympathetic modulation. Additionally, AI-generated calming tracks significantly promoted parasympathetic activation, evidenced by a median RMSSD increase of 2.48 () compared to the 0.79 increase observed in energizing tracks.
Stratifying participants based on daily music-listening duration (0--1, 2--3, and 4--5 hours) revealed high inter-individual variability, particularly among those with very low or high daily listening habits. These findings underscore the critical role of personal familiarity and preference in physiological responses, highlighting that AI-generated music may fall short in evoking the emotional resonance required for relaxation.
These mixed results point to the need for deeper personalization and dynamic adaptation to listeners’ preferences and physiological states. While statistical trends support the potential of targeted music interventions to positively influence autonomic regulation, challenges persist in achieving emotional authenticity and sustained HRV improvements.
Future research should focus on refining AI models for greater emotional depth, extending session durations, and integrating multimodal biometric data to optimize personalized, real-time interventions for emotional and physiological well-being
Alpina
Full text linkReports of mountaineering in the greater ranges in 2024 include a free climb of the central tower of the Torres del Paine in Patagonia, a first ascent of Hungchi in Nepal, toilet problems on Everest, a first ascent of Muchu Chhish in Pakistan, a difficult route on Wyoming’s Devils Tower, and a tribute to the late David Breashears, TM Herbert, and Lou Whittaker
News and Notes
Full text linkReflections on the December 18, 2023 flood caused a 6-inch rainstorm across northern Carroll and southern Coos Counties, New Hampshire. Appalachian Mountain Club trail builders create an All Persons Trail at AMC Cardigan Lodge in Alexandria, New Hampshire. Mountain author Mike Dickerman shares a photo memory of a bushwhack to Carrigain Pond in 1988
Books and Media
Full text linkReviews of: A Light Through the Cracks: A Climber\u27s Story, by Beth Rodden; Blood Sweat Tears, edited by Christine Reed; Pack Light: A Journey to Find Myself, by Shilletha Curtis; Alpine Rising: Sherpas, Baltis, and the Triumph of Local Climbers in the Greater Ranges, by Bernadette McDonald; Headstrap: Legends and Lore From the Climbing Sherpas of Darjeeling, by Nandini Purandare and Deepa Balsavar; Tramping Monadnock! New Discoveries with Henry David Thoreau, by Robert M. Young; Ghosts of Glencoe, by Chuck Schwerin; The Lifer: Rock Climbing Adventures in the Gunks and Beyond, by Russ Clune; Survival Is Not Assured: The Life of Climber Jim Donini, by Goeff Powter; Everest, Inc.: The Renegades and Rogues Who Built an Industry at the Top of the World, by Will Cockrell
Crowdsourced Data: Participatory Research and Community Knowledge
No full textData isn’t just gathered from remote sensors, administrative offices, or institutional surveys. Data can emerge dynamically from communities themselves. Crowdsourcing taps into collective participation, turning individuals into contributors who build and enhance datasets with lived experience and local knowledge. Examples range from apps like Waze and Google Maps, where drivers share real-time traffic data, to Harvard’s Crowd Counting Project, which verifies reports from participants and bystanders to count the size of public demonstrations and protests in the U.S., to the Red Dot Foundation’s SafeCity app, a crowdsourced platform providing information about sexual assault and harassment in public spaces. This exercise will highlight how crowdsourced data democratizes knowledge production and fosters community engagement. After exploring these examples, participants will break into small groups to brainstorm ideas for crowdsourcing data about Dartmouth’s own community, campus dynamics, and events
Rewiring Reality: The Neuroscience Behind Manifestation
No full textThis lesson introduces students to the scientific processes behind manifestation. Through interactive activities, students will learn to critically analyze claims in the media and create their own manifestation routines. They will leave this lesson with practical tools for well-being and transferable critical thinking skills
Your Words Matter
No full textThis lesson focuses on helping students understand the profound impact that affirmations and gratitude can have on their mindset and well-being. Applicable for everyone, students are encouraged to take their learnings and practice them in real life, hoping to promote a shift in perspective. Students will be able to recognize the benefits of their words and the way they speak to themselves, and understand how they can make changes in their lifestyle.
In this lesson, students will have a chance to reflect and learn about the science behind affirmations and gratitude while being introduced to strategies used to implement them in their everyday lives