Extensive data exploration for automatic price suggestion using item description: Case study for the Kaggle Mercari challenge

Amine Ait Si Ali; Huseyin Seker; Steven Farnie; John Elliott

doi:10.1145/3292448.3292458

Extensive data exploration for automatic price suggestion using item description: Case study for the Kaggle Mercari challenge

Amine Ait Si Ali, Huseyin Seker, Steven Farnie, John Elliott

Computer Information Sciences Department

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

This paper is related to a Kaggle competition. The competition is organised by Mercari and it is about building a model that can automatically and accurately suggests a selling price for a given item based on the information that the seller is providing. The provided information could be the description of the item, the category, the brand name, the item condition or the delivery option as well as other things. This is a regression problem and Natural Language Processing (NLP) techniques are used as well. An extensive data exploration is performed to help solving the problem. Logarithmic transformation is applied to skewed data and categorical features are combined with numerical ones. The developed modal produced promising results.

Original language	English
Title of host publication	ICAAI 2018 - 2018 the 2nd International Conference on Advances in Artificial Intelligence
Publisher	ACM
Pages	41-45
Number of pages	5
ISBN (Electronic)	9781450365833
DOIs	https://doi.org/10.1145/3292448.3292458
Publication status	Published - 6 Oct 2018
Event	2nd International Conference on Advances in Artificial Intelligence, ICAAI 2018 - Barcelona, Spain Duration: 6 Oct 2018 → 8 Oct 2018

Conference

Conference	2nd International Conference on Advances in Artificial Intelligence, ICAAI 2018
Country/Territory	Spain
City	Barcelona
Period	6/10/18 → 8/10/18

Keywords

Data visualisation
Feature union
Logaritmic transformation
Natural language processing
Price prediction
Regression

Access to Document

10.1145/3292448.3292458

Cite this

@inproceedings{fe78e700896f416da2c29460966cab80,

title = "Extensive data exploration for automatic price suggestion using item description: Case study for the Kaggle Mercari challenge",

abstract = "This paper is related to a Kaggle competition. The competition is organised by Mercari and it is about building a model that can automatically and accurately suggests a selling price for a given item based on the information that the seller is providing. The provided information could be the description of the item, the category, the brand name, the item condition or the delivery option as well as other things. This is a regression problem and Natural Language Processing (NLP) techniques are used as well. An extensive data exploration is performed to help solving the problem. Logarithmic transformation is applied to skewed data and categorical features are combined with numerical ones. The developed modal produced promising results.",

keywords = "Data visualisation, Feature union, Logaritmic transformation, Natural language processing, Price prediction, Regression",

author = "Ali, {Amine Ait Si} and Huseyin Seker and Steven Farnie and John Elliott",

year = "2018",

month = oct,

day = "6",

doi = "10.1145/3292448.3292458",

language = "English",

pages = "41--45",

booktitle = "ICAAI 2018 - 2018 the 2nd International Conference on Advances in Artificial Intelligence",

publisher = "ACM",

address = "United States",

note = "2nd International Conference on Advances in Artificial Intelligence, ICAAI 2018 ; Conference date: 06-10-2018 Through 08-10-2018",

}

Ali, AAS, Seker, H, Farnie, S & Elliott, J 2018, Extensive data exploration for automatic price suggestion using item description: Case study for the Kaggle Mercari challenge. in ICAAI 2018 - 2018 the 2nd International Conference on Advances in Artificial Intelligence. ACM, pp. 41-45, 2nd International Conference on Advances in Artificial Intelligence, ICAAI 2018, Barcelona, Spain, 6/10/18. https://doi.org/10.1145/3292448.3292458

Extensive data exploration for automatic price suggestion using item description: Case study for the Kaggle Mercari challenge. / Ali, Amine Ait Si; Seker, Huseyin; Farnie, Steven et al.
ICAAI 2018 - 2018 the 2nd International Conference on Advances in Artificial Intelligence. ACM, 2018. p. 41-45.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Extensive data exploration for automatic price suggestion using item description

T2 - 2nd International Conference on Advances in Artificial Intelligence, ICAAI 2018

AU - Ali, Amine Ait Si

AU - Seker, Huseyin

AU - Farnie, Steven

AU - Elliott, John

PY - 2018/10/6

Y1 - 2018/10/6

N2 - This paper is related to a Kaggle competition. The competition is organised by Mercari and it is about building a model that can automatically and accurately suggests a selling price for a given item based on the information that the seller is providing. The provided information could be the description of the item, the category, the brand name, the item condition or the delivery option as well as other things. This is a regression problem and Natural Language Processing (NLP) techniques are used as well. An extensive data exploration is performed to help solving the problem. Logarithmic transformation is applied to skewed data and categorical features are combined with numerical ones. The developed modal produced promising results.

AB - This paper is related to a Kaggle competition. The competition is organised by Mercari and it is about building a model that can automatically and accurately suggests a selling price for a given item based on the information that the seller is providing. The provided information could be the description of the item, the category, the brand name, the item condition or the delivery option as well as other things. This is a regression problem and Natural Language Processing (NLP) techniques are used as well. An extensive data exploration is performed to help solving the problem. Logarithmic transformation is applied to skewed data and categorical features are combined with numerical ones. The developed modal produced promising results.

KW - Data visualisation

KW - Feature union

KW - Logaritmic transformation

KW - Natural language processing

KW - Price prediction

KW - Regression

U2 - 10.1145/3292448.3292458

DO - 10.1145/3292448.3292458

M3 - Conference contribution

AN - SCOPUS:85061041449

SP - 41

EP - 45

BT - ICAAI 2018 - 2018 the 2nd International Conference on Advances in Artificial Intelligence

PB - ACM

Y2 - 6 October 2018 through 8 October 2018

ER -

Extensive data exploration for automatic price suggestion using item description: Case study for the Kaggle Mercari challenge

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this