A Comparative Study of Frequent Pattern Mining with Trajectory Data

Ding, Shiting; Li, Zhiheng; Zhang, Kai; Mao, Feng

doi:10.3390/s22197608

Cited by 5 publications

(4 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The study intends to use these approaches, specifically the DT algorithm, to forecast transportation modes from GPS trajectory data. This study contributes to the larger field of travel behavior analysis and provides legislators, trans-portation planners, and urban developers practical insights [9].…”

Section: Introductionmentioning

confidence: 95%

See 1 more Smart Citation

Machine Learning Algorithms for Transportation Mode Prediction: A Comparative Analysis

Murrar,

Alhaj,

Qutqut

2024

IJCAI

View full text Add to dashboard Cite

This study investigates the performance of various machine learning (ML) algorithms in predicting transportation modes from large datasets. The investigated algorithms include Multilayer Perceptron (MLP), K-Nearest Neighbors (KNN), Decision Tree (DT), Long Short-Term Memory (LSTM), Recurrent Neural Network (RNN), and Logistic Regression. We rigorously evaluated each algorithm's performance using a robust set of metrics such as precision, recall, and F1-score. This study comprehensively explains the algorithm's capabilities, strengths, and potential weaknesses across seven transportation categories: 'walk', 'bike ', 'bus', 'car', 'taxi', 'train', and 'subway'. The DT model consistently outperformed the others, demonstrating superior accuracy and an adequate balance of precision and recall across all modes of transportation. Specifically, it achieved precision, recall, and F1 scores of around 83% to 94% across all categories. These findings underline the suitability of the DT model for this classification task and its potential for further applications in transportation mode prediction based on large datasets. However, other algorithms like LSTM and RNN also showed promising results in certain categories, suggesting the value of continued exploration of different models depending on specific use cases.Povzetek: Raziskava preučuje učinkovitost algoritmov strojnega učenja pri napovedovanju načinov prevoza iz obsežnih podatkovnih zbirk, pri čemer izstopa model odločitvenih dreves.

show abstract

Section: Introductionmentioning

confidence: 95%

“…Various computational strategies have been developed over the years to deal with this difficulty. Among these, ML algorithms have emerged as particularly promising [9]. They can learn complex patterns from massive amounts of data, making them ideal for jobs such as transportation mode prediction.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Algorithms for Transportation Mode Prediction: A Comparative Analysis

Murrar,

Alhaj,

Qutqut

2024

IJCAI

View full text Add to dashboard Cite

show abstract

“…The study chose specific algorithms to analyse and explore representational sequence patterns. These algorithms were then assessed using different factors, with the most significant ones being the time it takes to run the algorithm and the amount of RAM it consumes [20].…”

Section: Related Workmentioning

confidence: 99%

Extraction of Association Rules from Cancer Patient’s Records using F-P Growth Algorithm

Alharith,

Khalil,

Ibrahim

et al. 2024

ITM Web Conf.

View full text Add to dashboard Cite

Cancer is a leading cause of mortality worldwide, and Sudan has a high cancer burden. The issue is that the data acquired from cancer patients grows yearly, and standard methodologies for analyzing this data are no longer adequate. Data mining techniques such as frequent pattern analysis and association rule mining are utilized in this research to assist in identifying hidden patterns and relationships in data. These strategies were utilized to provide valuable insights into the spread of cancer in Sudan and to assist healthcare professionals in making better diagnosis and treatment decisions. Support and confidence were utilized as measurement criteria. Support is used to evaluate the frequency of occurrence of an item or set of items among all transactions. In contrast, confidence is used to assess the strength of the relationship between groups of things. According to the findings, women are more likely than men to be diagnosed with cancer. The most common cancers in both genders include breast, prostate, ovarian, esophagus, and cervical cancers.

show abstract

“…Untuk melakukan identifikasi kawasan rawan kebakaran lahan di Kabupaten Kubu Raya, digunakan pemodelan matematik yaitu data mining dengan metode yang digunakan adalah metode sequential pattern mining. Sequential pattern mining merupakan penambangan serangkaian peristiwa atau urutan kejadian, dimana setiap urutan terdiri dari daftar elemen dan setiap elemen terdiri dari satu set item, dan diberi min_support yang ditentukan pengguna ambang batas, penambangan pola sekuensial adalah untuk menemukan urutan kejadian atau pola yang sering muncul, yaitu urutan yang frekuensi kemunculannya dalam rangkaian urutan tidak kurang dari min_support (Ding, Li, Zhang, & Mao, 2022) . Jikai nilai lift rasio lebih besar dari 1, maka kekuatan asosiasinya juga lebih besar.…”

Section: Analisis Dataunclassified

Sistem Informasi Geografis Pemetaan Kawasan Rawan Kebakaran Menggunakan Metode Sequential Pattern Mining

Melia¹,

Mutiah²,

Rahmayuda³

2023

j. inti nm

View full text Add to dashboard Cite

Land and forest fires occur every year in Indonesia, where the cause is human negligence or intentional related to forest deforestation. Land and forest fires, of course, can cause economic, health, social and cultural, ecological and environmental damage. One indicator that is useful as a determinant of the occurrence of land fires is hot spots. To manage hotspot data, data mining can be done, namely by using the sequential pattern mining method to obtain hotspot sequence patterns that will be used as indicators of land fires. In overcoming these problems, a Geographic Information System (GIS) was designed to map the areas prone to land fires in Kubu Raya Regency. The results of this study are maps of the distribution of land fire-prone areas and the pattern of occurrence of hotspots calculated using the sequential pattern mining method which is expected to assist relevant agencies in carrying out mitigation and prevention efforts for land fires, so as to produce the right decisions for handling land fires in the District. Kingdom Fortress. Based on the results of functionality testing using the black box testing method, the SIGPKRK application (Geographical Information System for Mapping Fire Prone Areas) that was built can run and is in accordance with the functions that have been designed. As for the results of testing the system interface using a questionnaire through the Google form which was filled in by 30 respondents, the SIGPKRK application obtained a percentage result of 85.9% which was included in the "Very Good" category.

show abstract

A Comparative Study of Frequent Pattern Mining with Trajectory Data

Cited by 5 publications

References 40 publications

Machine Learning Algorithms for Transportation Mode Prediction: A Comparative Analysis

Machine Learning Algorithms for Transportation Mode Prediction: A Comparative Analysis

Extraction of Association Rules from Cancer Patient’s Records using F-P Growth Algorithm

Sistem Informasi Geografis Pemetaan Kawasan Rawan Kebakaran Menggunakan Metode Sequential Pattern Mining

Contact Info

Product

Resources

About