Using SVM and Random forest for different features selection in predicting bike rental amount

Shiao, Yi Chen; Chung, Wei Hsiang; Chen, Rung Ching

doi:10.1109/icawst.2018.8517237

Cited by 7 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Meanwhile, city buses and mass-rapid transit (MRT) account for the majority of public transportation at 17% and 15%, respectively. The public transportation network in Taipei City was primarily established by connecting mid-and long-distance travel based on MRT and connecting short-distance travel using the public rental bicycle system after U-bike was launched in August 2012 in Taipei City (Shiao et al 2018;Chi et al 2019;Chang et al 2019). At present, U-bike has >12,379 bicycles operating out of 377 stations, which are installed at the exits of all MRT stations in Taipei City where bus stops are not fully covered.…”

Section: Research Hypothesis Method and Data Collectionmentioning

confidence: 99%

Characteristics of urban gardens and their accessibility to locals and non-locals in Taipei City, Taiwan

Hsiao

2021

Landscape Ecol Eng

View full text Add to dashboard Cite

The benefits and services of urban gardens (UGs) in urbanized areas are being increasingly recognized, leading to widespread enhancement of efforts in protection and provision. At present, several types of UGs created under the policy of "The Garden City Initiative" (GCI) exist in Taipei City, Taiwan, and four types of UGs, namely Allotment Garden (AG), Happy Garden (HG), Rooftop Garden (RG), and School Garden (SG), have been created using vacant lots and rooftop spaces developed via GCI.To clarify the characteristics of UGs in Taipei City, this paper first analyzes the changes in the number and surface area of each type of UG during 2015-2019. Second, using the 2019 data on all 201 UGs (18 AGs, 107 HGs, and 76 RGs) and excluding SGs that are only accessible by school students, this paper clarifies the spatial distribution of UGs in each district of Taipei City and the accessibility of UGs to locals and non-locals using a geographic information system. Furthermore, correlations between the accessibility of UGs to both locals and non-locals and their surface areas were evaluated. As a result, positive correlation was observed between the accessibility of UGs to locals and non-locals in seven districts, suggesting that both locals and non-locals have equal access to UGs.Lastly, the characteristics of UGs realized in Taipei City under GCI and their accessibility to locals and non-locals are discussed and conclusions are drawn on the implications for the utilization of vacant lots and rooftop spaces as UGs.

show abstract

Section: Research Hypothesis Method and Data Collectionmentioning

confidence: 99%

Characteristics of urban gardens and their accessibility to locals and non-locals in Taipei City, Taiwan

Hsiao

2021

Landscape Ecol Eng

View full text Add to dashboard Cite

show abstract

“…Başka bir çalışmada, Shiao vd. [6] özellik seçiminin tahmin yapma doğruluğu üzerindeki etkisinin, tahmin modelini tasarlama kadar çok olduğunu göstermişlerdir. Önişleme yapılmadan veri setinin kullanılması, yapılan tahmin sonucuna kötü yönde etki ettiği sonucuna varmışlardır.…”

Section: Giriş (Introduction)unclassified

Yarasa algoritması ile optimize edilmiş GBM modeli kullanarak mevsim bazlı bisiklet kiralama sayılarının tahmini

İleri

2024

Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi

View full text Add to dashboard Cite

Kentsel bisiklet talebinin etkili kaynak tahsisi için, paylaşımlı bisikletlerin doğru tahmin edilmesi gerekmektedir. Bu tahmin işlemi, Yarasa Algoritması (YA) ile optimize edilen Gradyan Artırmalı Makinesi (GBM) yöntemi kullanılarak gerçekleştirilmiştir. Önerilen modelin etkinliğini göstermek amacıyla, modelin performansı Karar Ağacı (DT), K-En Yakın Komşu (KNN) ve Çok Katmanlı Algılayıcı (MLP) gibi farklı yöntemlerle karşılaştırılmıştır. Bu karşılaştırma işlemi için MAE ve R2 metrikleri kullanılmıştır. En iyi sonuç 0.8780 R2 değerleri ile YA-GBM tarafından elde edilmiştir. Bununla birlikte, bisiklet kiralama sayısının tahminine en fazla ve en az etki eden özellikler de belirlenmiştir. En fazla etkiye sahip özellik hava sıcaklığı iken, en az etkiye sahip özellik ise kar yağışı olmuştur.

show abstract

Extracting Temporal-Based Spatial Features in Imbalanced Data for Predicting Dengue Virus Transmission

Setiyoutami

Anggraeni

Purwitasari

et al. 2020

Advances in Computer, Communication and Computational Sciences

View full text Add to dashboard Cite

Using SVM and Random forest for different features selection in predicting bike rental amount

Cited by 7 publications

References 6 publications

Characteristics of urban gardens and their accessibility to locals and non-locals in Taipei City, Taiwan

Characteristics of urban gardens and their accessibility to locals and non-locals in Taipei City, Taiwan

Yarasa algoritması ile optimize edilmiş GBM modeli kullanarak mevsim bazlı bisiklet kiralama sayılarının tahmini

Extracting Temporal-Based Spatial Features in Imbalanced Data for Predicting Dengue Virus Transmission

Contact Info

Product

Resources

About