Skip to main content
Journal of Big Data

Table 10 Average feature importance of the variables according to SHAP values

From: Predictive analytics using Big Data for the real estate market during the COVID-19 pandemic

Variable

Sell

Rent

May

June

July

August

May

June

July

August

Rooms

1.15580

0.49947

0.54986

0.55172

0.42647

0.35931

0.32831

0.46382

Sq_m_

1.60400

0.79154

0.77398

0.87373

0.44485

0.34983

0.21439

0.32318

Floor

0.52938

0.40655

0.37032

0.36983

0.17232

0.20210

0.20735

0.26256

Nr_Floors

0.65013

0.38285

0.46130

0.44349

0.18105

0.16328

0.35156

0.22225

TOM

2.47043

2.10286

1.98277

1.88522

1.24815

1.45630

1.16584

1.12961

Int_prices

2.28727

1.27575

1.30110

1.41305

1.26008

0.71070

0.62880

0.47666

Center

0.37411

0.30217

0.31365

0.37559

0.37768

0.41106

0.29141

0.31965

Year

–

1.54729

1.55409

1.48564

–

0.30344

0.29409

0.49686

Shop

–

0.76183

0.74752

0.94674

–

0.32995

0.38609

0.38636

School

–

0.74277

0.65204

0.65167

–

0.25219

0.31284

0.29746

Kinder

–

0.87993

0.75086

0.74460

–

0.28952

0.33553

0.37077

Furnish

–

0.20347

0.15533

0.16374

–

–

–

–

Heating

–

0.72057

0.66822

0.47049

–

0.40635

0.28597

0.47998

Built_type

–

–

1.13302

1.00620

–

–

0.57337

0.47470

  1. Bold values indicate the most dominant variable in predicting price change
Back to article page