Your privacy, your choice

We use essential cookies to make sure the site can function. We also use optional cookies for advertising, personalisation of content, usage analysis, and social media.

By accepting optional cookies, you consent to the processing of your personal data - including transfers to third parties. Some third parties are outside of the European Economic Area, with varying standards of data protection.

See our privacy policy for more information on the use of your personal data.

for further information and to change your choices.

You are viewing the site in preview mode

Skip to main content

Table 6 Comparison of Precision between SVR [31], MLP [32], RAQP [34], Vlachogianni (Vlacho) [4], LSTM [35], BLSTM [36], SLSTM [37], and the proposed BlaSt Models

From: Ecosense: a revolution in urban air quality forecasting for smart cities

APCs

Method

1hr

2hr

3hr

4hr

5hr

6hr

7hr

8hr

9hr

10hr

11hr

12hr

\(PM_{2.5}\)

SVR

0.938

0.904

0.892

0.862

0.802

0.798

0.753

0.735

0.710

0.697

0.655

0.633

 

MLP

0.958

0.927

0.895

0.856

0.823

0.790

0.770

0.758

0.733

0.714

0.710

0.703

 

RAQP

0.920

0.912

0.904

0.890

0.879

0.853

0.839

0.790

0.768

0.746

0.713

0.699

 

Vlacho

0.867

0.844

0.813

0.785

0.734

0.716

0.687

0.650

0.629

0.610

0.605

0.599

 

LSTM

0.948

0.923

0.919

0.893

0.863

0.833

0.813

0.793

0.753

0.740

0.725

0.711

 

BLSTM

0.966

0.920

0.890

0.865

0.843

0.825

0.815

0.793

0.773

0.743

0.723

0.703

 

SLSTM

0.979

0.954

0.934

0.913

0.880

0.863

0.825

0.801

0.790

0.765

0.743

0.723

 

BlaSt

0.995

0.975

0.930

0.910

0.896

0.892

0.880

0.860

0.860

0.850

0.840

0.828

\(NO_{2}\)

SVR

0.915

0.913

0.903

0.888

0.883

0.873

0.863

0.857

0.843

0.798

0.783

0.778

 

MLP

0.900

0.878

0.838

0.832

0.808

0.783

0.760

0.753

0.724

0.720

0.703

0.701

 

RAQP

0.966

0.958

0.928

0.910

0.884

0.867

0.839

0.828

0.798

0.763

0.748

0.707

 

Vlacho

0.868

0.848

0.840

0.835

0.823

0.802

0.797

0.790

0.720

0.709

0.705

0.703

 

LSTM

0.963

0.943

0.940

0.905

0.884

0.856

0.847

0.784

0.760

0.751

0.737

0.718

 

BLSTM

0.975

0.954

0.930

0.905

0.884

0.857

0.837

0.800

0.789

0.760

0.746

0.718

 

SLSTM

0.980

0.978

0.950

0.925

0.883

0.856

0.837

0.810

0.790

0.772

0.747

0.728

 

BlaSt

0.992

0.980

0.903

0.889

0.884

0.874

0.865

0.858

0.848

0.843

0.838

0.836

CO

SVR

0.923

0.889

0.869

0.854

0.834

0.808

0.778

0.762

0.720

0.697

0.688

0.678

 

MLP

0.952

0.946

0.888

0.858

0.828

0.798

0.778

0.760

0.739

0.730

0.712

0.706

 

RAQP

0.960

0.933

0.898

0.898

0.893

0.873

0.869

0.865

0.852

0.843

0.820

0.778

 

Vlacho

0.872

0.867

0.863

0.860

0.859

0.857

0.849

0.841

0.813

0.804

0.790

0.784

 

LSTM

0.973

0.934

0.900

0.875

0.843

0.820

0.787

0.750

0.738

0.722

0.717

0.703

 

BLSTM

0.989

0.954

0.930

0.905

0.884

0.856

0.817

0.800

0.780

0.760

0.737

0.708

 

SLSTM

0.980

0.954

0.933

0.908

0.864

0.839

0.817

0.790

0.760

0.730

0.717

0.701

 

BlaSt

0.995

0.965

0.925

0.897

0.896

0.886

0.878

0.865

0.857

0.847

0.836

0.834

  1. Bold values indicate the efficiency as well as the superior performance of our proposed model
Back to article page

BMC Research Notes

ISSN: 1756-0500

Contact us