Automatic Generation of Natural Language Explanations

Felipe Soares Da Costa; Sixun Ouyang; Peter Dolog; Aonghus Lawlor

doi:10.1145/3180308.3180366

Automatic Generation of Natural Language Explanations

Felipe Soares Da Costa, Sixun Ouyang, Peter Dolog, Aonghus Lawlor

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

54 Citations (Scopus)

Abstract

An interesting challenge for explainable recommender systems is to provide successful interpretation of recommendations using structured sentences. It is well known that user-generated reviews, have strong influence on the users' decision. Recent techniques exploit user reviews to generate natural language explanations. In this paper, we propose a character-level attention-enhanced long short-term memory model to generate natural language explanations. We empirically evaluated this network using two real-world review datasets. The generated text present readable and similar to a real user's writing, due to the ability of reproducing negation, misspellings, and domain-specific vocabulary.

Original language	English
Title of host publication	Proceedings of the 23rd International Conference on Intelligent User Interfaces
Publisher	Association for Computing Machinery
Publication date	8 Mar 2018
Article number	57
ISBN (Electronic)	978-1-4503-5571-1
DOIs	https://doi.org/10.1145/3180308.3180366
Publication status	Published - 8 Mar 2018
Event	International Conference on Intelligent User Interfaces - Hitotsubashi Hall (National Center of Sciences Building), Tokyo, Japan Duration: 7 Mar 2018 → 11 Mar 2018 Conference number: 23 https://iui.acm.org/2018/

Conference

Conference	International Conference on Intelligent User Interfaces
Number	23
Location	Hitotsubashi Hall (National Center of Sciences Building)
Country/Territory	Japan
City	Tokyo
Period	07/03/2018 → 11/03/2018
Internet address	https://iui.acm.org/2018/

Keywords

Explainability
Explanations
Natural language generation
Neural network
Recommender systems

Access to Document

10.1145/3180308.3180366

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

@inproceedings{672eba8660dc42f78fc745d38271a22b,

title = "Automatic Generation of Natural Language Explanations",

abstract = "An interesting challenge for explainable recommender systems is to provide successful interpretation of recommendations using structured sentences. It is well known that user-generated reviews, have strong influence on the users' decision. Recent techniques exploit user reviews to generate natural language explanations. In this paper, we propose a character-level attention-enhanced long short-term memory model to generate natural language explanations. We empirically evaluated this network using two real-world review datasets. The generated text present readable and similar to a real user's writing, due to the ability of reproducing negation, misspellings, and domain-specific vocabulary.",

keywords = "Explainability, Explanations, Natural language generation, Neural network, Recommender systems",

author = "{Da Costa}, {Felipe Soares} and Sixun Ouyang and Peter Dolog and Aonghus Lawlor",

year = "2018",

month = mar,

day = "8",

doi = "10.1145/3180308.3180366",

language = "English",

booktitle = "Proceedings of the 23rd International Conference on Intelligent User Interfaces",

publisher = "Association for Computing Machinery",

address = "United States",

note = "International Conference on Intelligent User Interfaces, IUI ; Conference date: 07-03-2018 Through 11-03-2018",

url = "https://iui.acm.org/2018/",

}

Automatic Generation of Natural Language Explanations. / Da Costa, Felipe Soares; Ouyang, Sixun; Dolog, Peter et al.
Proceedings of the 23rd International Conference on Intelligent User Interfaces. Association for Computing Machinery, 2018. 57.

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Automatic Generation of Natural Language Explanations

AU - Da Costa, Felipe Soares

AU - Ouyang, Sixun

AU - Dolog, Peter

AU - Lawlor, Aonghus

N1 - Conference code: 23

PY - 2018/3/8

Y1 - 2018/3/8

N2 - An interesting challenge for explainable recommender systems is to provide successful interpretation of recommendations using structured sentences. It is well known that user-generated reviews, have strong influence on the users' decision. Recent techniques exploit user reviews to generate natural language explanations. In this paper, we propose a character-level attention-enhanced long short-term memory model to generate natural language explanations. We empirically evaluated this network using two real-world review datasets. The generated text present readable and similar to a real user's writing, due to the ability of reproducing negation, misspellings, and domain-specific vocabulary.

AB - An interesting challenge for explainable recommender systems is to provide successful interpretation of recommendations using structured sentences. It is well known that user-generated reviews, have strong influence on the users' decision. Recent techniques exploit user reviews to generate natural language explanations. In this paper, we propose a character-level attention-enhanced long short-term memory model to generate natural language explanations. We empirically evaluated this network using two real-world review datasets. The generated text present readable and similar to a real user's writing, due to the ability of reproducing negation, misspellings, and domain-specific vocabulary.

KW - Explainability

KW - Explanations

KW - Natural language generation

KW - Neural network

KW - Recommender systems

UR - http://www.scopus.com/inward/record.url?scp=85045131728&partnerID=8YFLogxK

U2 - 10.1145/3180308.3180366

DO - 10.1145/3180308.3180366

M3 - Article in proceeding

BT - Proceedings of the 23rd International Conference on Intelligent User Interfaces

PB - Association for Computing Machinery

T2 - International Conference on Intelligent User Interfaces

Y2 - 7 March 2018 through 11 March 2018

ER -

Automatic Generation of Natural Language Explanations

Abstract

Conference

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Cite this