{"id":31006417,"url":"https://github.com/anthonylucky1909/dtwboost","last_synced_at":"2026-04-20T23:32:48.205Z","repository":{"id":190440499,"uuid":"682627964","full_name":"anthonylucky1909/DTWBoost","owner":"anthonylucky1909","description":"DTWBoost: A Dynamic Time Warping Enhanced SVM Classifier for Time Series Data","archived":false,"fork":false,"pushed_at":"2025-08-21T00:21:05.000Z","size":21117,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-12-30T07:30:32.634Z","etag":null,"topics":["artificial-intelligence","dtw-algorithm","svm-classifier","time-series"],"latest_commit_sha":null,"homepage":"","language":"TypeScript","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/anthonylucky1909.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":"2023-08-24T15:19:09.000Z","updated_at":"2025-09-23T01:59:25.000Z","dependencies_parsed_at":null,"dependency_job_id":"5ab20dfb-82e7-4265-a9dc-f230857b8ca6","html_url":"https://github.com/anthonylucky1909/DTWBoost","commit_stats":null,"previous_names":["antshuang19/evaluation-kernel-using-timeserie-dataset","antshuang19/dtw-kernel-classifier","anthonyhuang19/dtw-svm-x","anthonylucky1909/dtw-svm-x","anthonylucky1909/dtwboost"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/anthonylucky1909/DTWBoost","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anthonylucky1909%2FDTWBoost","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anthonylucky1909%2FDTWBoost/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anthonylucky1909%2FDTWBoost/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anthonylucky1909%2FDTWBoost/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/anthonylucky1909","download_url":"https://codeload.github.com/anthonylucky1909/DTWBoost/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anthonylucky1909%2FDTWBoost/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32070656,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-20T21:26:33.338Z","status":"ssl_error","status_checked_at":"2026-04-20T21:26:22.081Z","response_time":94,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":["artificial-intelligence","dtw-algorithm","svm-classifier","time-series"],"created_at":"2025-09-13T02:11:47.951Z","updated_at":"2026-04-20T23:32:48.174Z","avatar_url":"https://github.com/anthonylucky1909.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# DTWBoost: A Dynamic Time Warping Enhanced SVM Classifier for Time Series Data\n\n[![PyPI version](https://img.shields.io/pypi/v/dtwboost.svg)](https://pypi.org/project/dtwboost/)\n[![PyPI downloads](https://img.shields.io/pypi/dm/dtwboost.svg)](https://pypi.org/project/dtwboost/)\n[![Python versions](https://img.shields.io/pypi/pyversions/dtwboost.svg)](https://www.python.org/)\n[![Open issues](https://img.shields.io/github/issues/huangjunxiang/DTWBoost.svg)](https://github.com/huangjunxiang/DTWBoost/issues)\n[![License](https://img.shields.io/github/license/huangjunxiang/DTWBoost.svg)](https://github.com/huangjunxiang/DTWBoost/blob/main/LICENSE)\n[![Coverage](https://codecov.io/gh/huangjunxiang/DTWBoost/branch/main/graph/badge.svg)](https://codecov.io/gh/huangjunxiang/DTWBoost)\n\n---\n\n## 📌 Abstract\nThis project evaluates the use of **Support Vector Machine (SVM)** with different **kernel functions** combined with **Dynamic Time Warping (DTW)** for classifying time series data.  \nWe assess performance on datasets such as **ECG**, **FordA**, and **HAR**.\n\n---\n\n## 🔑 Keywords\n- Support Vector Machine (SVM)  \n- Dynamic Time Warping (DTW)  \n- Kernel Functions  \n- Time Series Classification  \n\n---\n\n## 🎯 Objective\nEvaluate the effectiveness of various DTW-based SVM kernels on different time series datasets.\n\n---\n\n## 📚 Related Work\nPast research explored DTW-SVM combinations. Kernels like Gaussian, Polynomial, and custom DTW-based ones have shown varying success.\n\n---\n\n## ⚙️ Methodology\n\n### Steps\n1. **Preprocessing** – Normalize and balance data  \n2. **DTW Distance Calculation**  \n3. **Kernel Application** – Replace Euclidean distance with DTW in kernel  \n4. **SVM Training** – With cross-validation  \n5. **Evaluation** – Accuracy and confusion matrix  \n\n### Kernels Used\n- **Cauchy**  \n- **Gaussian**  \n- **Inverse Multiquadric**  \n- **Laplacian**  \n- **Logarithmic**  \n- **Rational Quadratic**  \n\n---\n\n## 📊 Datasets\n- **ECG** – Heartbeat signal classification  \n- **FordA** – Engine fault detection  \n- **HAR** – Human activity recognition via smartphone sensors  \n\n---\n\n## 🔧 Experimental Settings\n\n### Metrics\n- **Accuracy**  \n- **Confusion Matrix**\n\n### Sample Results (ECG)\n| Kernel       | Hyperparameter | Time (s) |\n|--------------|----------------|----------|\n| Cauchy       | sigma=500000   | 213.1    |\n| Gaussian     | sigma=1        | 310.7    |\n| Inverse MQ   | c=0.4          | 346.7    |\n\n### Accuracy Comparison\n| Kernel            | ECG (%) | FordA (%) | HAR (%) |\n|-------------------|---------|-----------|---------|\n| Cauchy            | 89.0    | 60.5      | 70.0    |\n| Gaussian          | 76.0    | 45.0      | 57.1    |\n| Inverse MQ        | 85.0    | 60.0      | 80.3    |\n| Laplacian         | 83.0    | 53.5      | 82.7    |\n| Logarithmic       | 38.0    | 53.0      | 85.6    |\n| Rational Quadratic| 84.0    | 54.0      | 87.0    |\n\n---\n\n## 💬 Insights\n- **Inverse Multiquadric** performs best across datasets.  \n- DTW improves SVM accuracy but increases computation time.  \n- FordA is harder due to signal variance.  \n\n---\n\n## 🧠 Discussion\n- **DTW** is effective but slow; use **FastDTW** or parallelism for scalability.  \n- DTW kernels help with noisy, irregular-length time series.  \n\n---\n\n## ✅ Conclusion\nCombining DTW with flexible SVM kernels significantly improves time series classification.  \n**Rational Quadratic** and **Inverse Multiquadric** performed best overall.  \n\n---\n\n## 📖 References\n1. Lei \u0026 Sun, *Dynamic Time Warping in Kernel Machines*  \n2. Hofmann et al., *Kernel Methods in ML*  \n3. Zhang, *List of SVM Kernels*  \n4. Reyes-Ortiz, *HAR Dataset*  \n5. Bagnall, *FordA Dataset*  \n6. Kachuee et al., *ECG Classification*  \n7. [NCBI DTW Paper](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4120293/)  \n\n---\n\n\u003c!-- Badge References --\u003e\n[pypi-badge]: https://img.shields.io/pypi/v/dtwboost.svg\n[pypi-url]: https://pypi.org/project/dtwboost/\n[downloads-badge]: https://img.shields.io/pypi/dm/dtwboost.svg\n[downloads-url]: https://pypi.org/project/dtwboost/\n[python-badge]: https://img.shields.io/pypi/pyversions/dtwboost.svg\n[python-url]: https://www.python.org/\n[issues-badge]: https://img.shields.io/github/issues/huangjunxiang/DTWBoost.svg\n[issues-url]: https://github.com/huangjunxiang/DTWBoost/issues\n[license-badge]: https://img.shields.io/github/license/huangjunxiang/DTWBoost.svg\n[license-url]: https://github.com/huangjunxiang/DTWBoost/blob/main/LICENSE\n[coverage-badge]: https://codecov.io/gh/huangjunxiang/DTWBoost/branch/main/graph/badge.svg\n[coverage-url]: https://codecov.io/gh/huangjunxiang/DTWBoost\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fanthonylucky1909%2Fdtwboost","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fanthonylucky1909%2Fdtwboost","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fanthonylucky1909%2Fdtwboost/lists"}