A machine learning study of 2,500 synthetic adolescent profiles — exploring how stress, sleep, physical activity, social media use, and everyday habits relate to depression risk, with an interactive model you can explore for yourself.
A browser version of the study's interactive tool, driven by the logistic model's published coefficients (approximated for the web). Inputs are grouped by role — adjust them to see how the model's estimate responds, and notice how much stress and sleep move it compared to everything else. This is a toy built on made-up data, meant to show how a machine learning model "thinks" — it says nothing about any real person, including you.
This is not a screening or diagnostic tool, and not a substitute for talking to a professional about mental health.
"Lowkey stressed." "Lowkey tired." It's how this generation talks about struggle — quiet words for things that aren't quiet at all. LOWKEY is a data-driven mental health project that takes those quiet words seriously: a machine learning study of what actually signals depression risk in adolescents, paired with honest evidence, practical guidance, and clear paths to help.
Move the conversation beyond screen-time blame, toward what the evidence shows matters most — and make asking for help feel as normal as asking "kamusta ka?"
A generation where no young person struggles quietly, and no adult around them is afraid to ask.
Each record represents one adolescent: demographics, social-media habits, sleep, academic performance, physical activity, social interaction, and self-reported stress and anxiety. The target is depression_risk — a three-class label (low / medium / high). Categorical columns were integer-encoded to make the table AI-ready.
Low risk is the majority class (50.6%) — one reason macro-averaged F1 was used, so every class counts equally.
First five records from the raw survey table.
| Age | Media h/day | Platform | Sleep h | Stress | Anxiety | Risk |
|---|---|---|---|---|---|---|
| 19 | 3.8 | 6.5 | 6 | 6 | medium | |
| 16 | 6.5 | Other | 5.8 | 9 | 7 | high |
| 17 | 2.9 | Other | 7.0 | 3 | 2 | low |
| 19 | 7.9 | TikTok | 5.0 | 10 | 8 | medium |
| 15 | 1.2 | TikTok | 8.8 | 1 | 2 | low |
Logistic Regression (L1 vs L2, six values of C, 10 trials each), then Random Forest and Gradient Boosting with small hyperparameter searches. All compared on mean testing macro F1 across stratified splits.
Testing macro F1 vs regularization strength C (log scale). L1 at C = 0.1 wins the baseline round.
Mean testing scores per model. Gradient Boosting takes every metric.
C = 0.1
300 trees · max depth 5
200 estimators · learning rate 0.01
Data. 2,500 synthetic adolescent records, 11 predictors, target depression_risk ∈ {low, medium, high} (1,266 / 650 / 584 — imbalanced, hence macro-averaged metrics that weight every class equally). Categorical variables integer-encoded; all features standardized with StandardScaler fit on training data only.
Evaluation protocol. 75/25 train–test splits, stratified on the target; every hyperparameter configuration evaluated over 10 trials with varying random seeds, selecting by mean testing macro F1. Final models refit with random_state=42.
Search spaces. Logistic Regression (liblinear, one-vs-rest): penalty ∈ {L1, L2} × C ∈ {0.001, 0.01, 0.1, 1, 10, 100} → best L1, C=0.1 (F1 0.666). Random Forest: trees ∈ {100, 200, 300} × depth ∈ {None, 5, 10} → best 300 trees, depth 5 (F1 0.759). Gradient Boosting: estimators ∈ {100, 200} × learning rate ∈ {0.01, 0.1, 0.2} → best 200, lr=0.01 (F1 0.774, accuracy 0.799).
Convergent feature importance. Gradient Boosting's impurity-based importances and L1's sparsity agree: L1 zeroed out 5 of 11 features and retained stress (mean |coef| 2.05) and sleep (0.24) — an implicit feature-selection check that the 93% stress finding isn't an artifact of one model family.
Browser demo disclosure. The interactive tool above runs the L1 model's published per-class coefficients in JavaScript, with feature standardization statistics estimated from the data's plausible ranges and intercepts calibrated to reproduce the training class balance — a faithful qualitative approximation, not the pickled model. Exact parity would require exporting the fitted sklearn pipeline (e.g., to ONNX). Stated limitations: synthetic data, correlational design, self-reported measures, and no external validation cohort.
Read this before the number below: one feature explaining 93% of a model's predictions is not a typical finding in real behavioral research — it's the signature of a synthetic dataset whose labels were likely generated with a formula that weighted stress heavily. Treat the 93% as a demonstration of how feature-importance analysis works, not as evidence about real teenagers. The peer-reviewed studies below are where the real evidence lives.
Both of the best-performing models converge on the same pattern within this synthetic dataset: self-reported stress level accounted for the large majority of predictive information — 93% of Gradient Boosting's feature importance — with sleep duration a consistent second. Time spent on social media contributed comparatively little once stress was accounted for. This pattern is at least directionally consistent with real clinical understanding — sustained stress and disrupted sleep are well-established correlates of depressive symptoms in adolescents — but the specific 93% figure is a property of this dataset, not a real-world measurement.
Share of total importance (log-scaled axis so the small contributors stay visible).
Three takeaways from 2,500 profiles and three model families — and what they suggest for the adults and teens reading this.
Stress level alone accounted for 93% of the model's predictive information. For parents and teachers, the practical takeaway is about where attention goes: rather than starting with "how long were you online?", start with "what's been weighing on you lately?" — and make space to listen to the answer.
Sleep duration ranked a consistent second across models — and it's the input that tips borderline cases in the tool above. Adolescents need roughly 8–10 hours of sleep; protecting a regular sleep routine is one of the most concrete, evidence-aligned things a household can support.
Social-media hours, platform choice, and screen use before bed together contributed less than 2% of the signal. Heavy screen use may be how distress looks from the outside — a way of coping or withdrawing — so restricting devices alone can address the visible behavior while leaving the underlying need unmet.
How to read these results responsibly
This study uses synthetic classroom data, and association is not causation — the model describes what tends to occur together, not what causes what. In particular, nothing here implies that low grades cause depression: academic changes are usually a visible consequence of distress, not its source. A falling grade is a reason to ask "kamusta ka?" — never a reason to add pressure. No model output can assess an individual adolescent. Depression is diagnosable and treatable, and concerns about a young person's mood, sleep, withdrawal, or functioning are best brought to a physician, psychologist, or school counselor.
One peer-reviewed study per factor — chosen for scale and rigor, not for making the strongest headline:
Notice how this pattern echoes the study's own finding: the two factors with the deepest, most consistent evidence base — stress and sleep — are also the two that carried this model's signal, while academic performance and social media, despite real associations in the literature, contributed comparatively little here.
This is the actual PHQ-9 screening questionnaire adapted for adolescents by the GLAD-PC team, hosted by the American Academy of Child & Adolescent Psychiatry. Unlike the demo above, this is a real instrument used in real clinics. In clinics, teens fill this out themselves and a clinician reviews it with them — that's exactly how it's meant to be used here too.
Before you start: this is a screening tool, not a diagnosis — only a qualified professional can diagnose depression. Nothing you answer here is saved, stored, or sent anywhere — everything happens in your browser and disappears when you close or refresh the page. Your answers are also never used to train, retrain, or improve the machine learning model on this site — the two are completely separate. The most useful thing to do with your results is show them to a parent, doctor, or school counselor.
In the past year, have you felt depressed or sad most days, even if you felt okay sometimes?
How difficult have these problems made it for you to do your work, take care of things at home, or get along with other people?
Has there been a time in the past month when you have had serious thoughts about ending your life?
Have you EVER, in your WHOLE LIFE, tried to kill yourself or made a suicide attempt?
Source: Modified with permission by the GLAD-PC team from the PHQ-9 (Spitzer, Williams & Kroenke, 1999), Revised PHQ-A (Johnson, 2002), and the CDS (DISC Development Group, 2000). Guidelines for Adolescent Depression in Primary Care, Version 2, 2010. Hosted by AACAP: aacap.org (PDF).
Beyond the everyday habits covered in the model above, here's how to support a friend in the moment, and where to find real help in the Philippines. (Looking for a screening tool? Try the PHQ-9 Screening section.)
Most teens tell a friend before they ever tell an adult — which means you might be the first to know. Two reassurances before the scripts: needing help is normal, and asking for it is normal — for your friend, and for you if the weight gets heavy. You don't need perfect words. These work:
To open the door: "Napansin ko lang, parang iba ka these days. Kamusta ka?" · "You don't have to explain everything. I just want you to know I noticed."
To keep it open: "That sounds really heavy. I'm glad you told me." · "Gusto mo bang samahan kita to talk to the guidance counselor? We can go together."
Two things to remember: listening is not the same as fixing — you don't have to solve it. And if a friend talks about hurting themselves, that's never a secret to keep, even if they ask you to. Telling a trusted adult or calling a hotline together is how you stay a good friend, not how you betray one.
When concerns go beyond what a check-in can hold, these professional societies maintain directories of qualified specialists across the Philippines:
Free, confidential, and staffed by trained responders. Save this number in your phone.
More avenues of support: your school's guidance office, your barangay health center, and the National Center for Mental Health (ncmh.gov.ph). Under the Philippine Mental Health Act (RA 11036), access to mental health services is a legal right. Number verified as of July 2026.
LOWKEY is built to align with the Philippine Data Privacy Act of 2012 (RA 10173) and the guidance of the National Privacy Commission (NPC) — even though, as explained below, this site is designed to collect almost nothing in the first place. Privacy by design means the simplest, safest choice: don't collect what you don't need.
Two small, unavoidable exceptions, both routine for any website or offline-capable app:
Because this project doesn't collect, store, or process personal data on a server, it doesn't currently need to register as a Personal Information Controller with the NPC — but it follows RA 10173's core principles anyway: transparency (this notice), legitimate purpose (education, not data collection), and proportionality (collect nothing beyond what's needed — which, here, is nothing at all).
Even though this site holds no personal data of yours to act on, RA 10173 guarantees every data subject the right to: be informed, access their data, object to processing, request correction or erasure, data portability, and to file a complaint with the National Privacy Commission (privacy.gov.ph). If that ever changes — for example, if a future version of this project collects data for research, as described in the Improve section below — it would only do so with your explicit, informed consent, gathered separately from simply browsing this page.
LOWKEY is a class project, not a finished product — and it should say so. Here's what we know is missing or approximate, in the interest of being honest about its limits.
Have a suggestion?
This project improves through feedback. If something here felt wrong, unclear, or missing — especially from teens, parents, or mental health professionals — that input is exactly what the next version needs.
Sending opens your email app with everything pre-filled, addressed to the project team — nothing is sent from this page directly, and no account is needed. This is for feedback about the project only; for urgent concerns, use the hotline above instead.
THE REAL CALL TO ACTION
Sometimes the most helpful response is asking, gently and without judgment,
“kamusta ka?”
Check in on the young people in your life. Ask about stress, listen more than you advise, and protect their sleep — the data, and clinical experience, agree that both matter.
If you or someone you know is struggling, help is available and reaching out is a sign of strength — talk to a physician, mental health professional, school counselor, or a trusted adult, or call the hotlines listed above, including the 24/7 NCMH Crisis Hotline at 1553.