{"id":31642238,"url":"https://github.com/belpa03/mobile-device-usage-user-behavior","last_synced_at":"2026-05-10T07:47:25.983Z","repository":{"id":318128707,"uuid":"1070087435","full_name":"belpa03/mobile-device-usage-user-behavior","owner":"belpa03","description":"Deep learning-based analysis of mobile device usage and user behavior using ANN. Includes data cleaning, EDA, feature engineering, model training, and infographic visualization.","archived":false,"fork":false,"pushed_at":"2025-10-05T08:55:29.000Z","size":780,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-10-05T10:27:33.293Z","etag":null,"topics":["ann","deep-learning","eda","keras","phyton","tensorflow","user-behavior"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/belpa03.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-10-05T08:35:56.000Z","updated_at":"2025-10-05T08:55:32.000Z","dependencies_parsed_at":"2025-10-05T10:27:38.954Z","dependency_job_id":null,"html_url":"https://github.com/belpa03/mobile-device-usage-user-behavior","commit_stats":null,"previous_names":["belpa03/mobile-device-usage-user-behavior"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/belpa03/mobile-device-usage-user-behavior","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/belpa03%2Fmobile-device-usage-user-behavior","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/belpa03%2Fmobile-device-usage-user-behavior/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/belpa03%2Fmobile-device-usage-user-behavior/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/belpa03%2Fmobile-device-usage-user-behavior/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/belpa03","download_url":"https://codeload.github.com/belpa03/mobile-device-usage-user-behavior/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/belpa03%2Fmobile-device-usage-user-behavior/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":278717450,"owners_count":26033542,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-10-07T02:00:06.786Z","response_time":59,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ann","deep-learning","eda","keras","phyton","tensorflow","user-behavior"],"created_at":"2025-10-07T03:57:27.398Z","updated_at":"2025-10-07T03:57:34.692Z","avatar_url":"https://github.com/belpa03.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 📱 Mobile Device Usage \u0026 User Behavior — ANN Deep Learning Project\n---\n\n## 🔍 Dataset Overview\n**Source:** [Kaggle - Mobile Device Usage \u0026 User Behavior Dataset](https://www.kaggle.com/code/momenayman9/mobile-device-usage-and-user-behavio/input)\n\n| Variable | Description |\n|----------|-------------|\n| User ID | Unique identifier for each user |\n| Device Model | Smartphone model used |\n| Operating System | Device OS (iOS or Android) |\n| App Usage Time (min/day) | Average daily app usage in minutes |\n| Screen On Time (hours/day) | Average daily screen-on duration |\n| Battery Drain (mAh/day) | Daily battery consumption |\n| Number of Apps Installed | Total installed apps |\n| Data Usage (MB/day) | Daily data usage in MB |\n| Age | User age |\n| Gender | Male / Female |\n| User Behavior Class | Behavior classification (1–5, light to extreme) |\n\n---\n\n## 🔁 Project Workflow\n\n### 1. Data Ingestion \u0026 Initial Check\n- Load dataset and inspect schema.\n- Check for missing values and inconsistent types.\n- No missing values found → ready for preprocessing.\n\n### 2. Exploratory Data Analysis (EDA)\n- Analyze distributions of User Behavior Class, Age, Gender, OS.\n- Visualize correlations between App Usage Time, Screen On Time, Battery Drain, Data Usage.\n- Insights: heavy usage correlates with high battery consumption; Class 2 dominates.\n\n### 3. Data Cleaning \u0026 Preprocessing\n- Encode categorical variables.\n- Standardize numeric variables for ANN modeling.\n- Dataset split into 80% training / 20% testing.\n\n### 4. Feature Engineering\n- Create derived metrics (e.g., usage per app, battery per hour).\n- Correlation analysis to select features for ANN.\n\n### 5. ANN Model Training\n- Train ANN with training set, validation split 0.33, 100 epochs.\n- Monitor accuracy and loss to ensure proper convergence.\n\n### 6. Model Evaluation\n- Evaluate test accuracy and analyze confusion matrix.\n- Class 1 shows highest correct-classification rate (73.8%).\n\n### 7. Predictions\n- Generate predicted classes for test set.\n- Evaluate probability distributions across classes.\n\n### 8. Model Visualization\n- Infographic charts for training accuracy, loss, and confusion matrix.\n- Heatmaps and scatterplots for feature correlations.\n\n### 9. Reporting \u0026 Business Insights\n- Infographic + one-page summary for stakeholders.\n- Provides actionable insights for app retention, device optimization, and personalization strategies.\n\n---\n\n## 💡 Conclusions\n\n- Device usage intensity directly impacts battery consumption and app interaction.  \n- Behavior Class 2 dominates the population, while Class 1 is least represented.  \n- ANN achieved **76.62% accuracy**, showing good predictive performance.  \n- Confusion matrix shows Class 1 was most accurately predicted (73.8%).  \n- Misclassifications mainly occur between adjacent classes (2 and 3), suggesting overlapping behavioral traits.  \n- ANN effectively models nonlinear relationships, supporting segmentation and targeted interventions for different user behavior classes.  \n\n---\n\n## 🖼️ Infographic Notes (`User Behavior.png`)\n![User Behavior](User%20Behavior.jpeg)  \n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbelpa03%2Fmobile-device-usage-user-behavior","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbelpa03%2Fmobile-device-usage-user-behavior","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbelpa03%2Fmobile-device-usage-user-behavior/lists"}