Towards Recommender Systems for Police Photo Lineup

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11210%2F17%3A10362647" target="_blank" >RIV/00216208:11210/17:10362647 - isvavai.cz</a>

Alternative codes found

RIV/00216208:11320/17:10362647

Result on the web

<a href="http://dx.doi.org/10.1145/3125486.3125490" target="_blank" >http://dx.doi.org/10.1145/3125486.3125490</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1145/3125486.3125490" target="_blank" >10.1145/3125486.3125490</a>

Result language

angličtina

Original language name

Towards Recommender Systems for Police Photo Lineup

Original language description

Photo lineups play a significant role in the eyewitness identifica-tion process. This method is used to provide evidence in the prosecution and subsequent conviction of suspects. Unfortu-nately, there are many cases where lineups have led to the con-viction of an innocent suspect. One of the key factors affecting the incorrect identification of a suspect is the lack of lineup fair-ness, i.e. that the suspect differs significantly from all other candidates. Although the process of assembling fair lineup is both highly important and time-consuming, only a handful of tools are available to simplify the task. In this paper, we describe our work towards using recommend-er systems for the photo lineup assembling task. We propose and evaluate two complementary methods for item-based rec-ommendation: one based on the visual descriptors of the deep neural network, the other based on the content-based attrib-utes of persons. The initial evaluation made by forensic technicians shows that although results favored visual descriptors over attribute-based similarity, both approaches are functional and highly diverse in terms of recommended objects. Thus, future work should in-volve incorporating both approaches in a single prediction method, preference learning based on the feedback from forensic technicians and recommendation of assembled lineups instead of single candidates.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

<a href="/en/project/GA17-22224S" target="_blank" >GA17-22224S: User preference analytics in multimedia exploration models</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2017

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Proceedings of the 2nd Workshop on Deep Learning for Recommender Systems

ISBN

978-1-4503-5353-3

ISSN

—

e-ISSN

neuvedeno

Number of pages

5

Pages from-to

19-23

Publisher name

Association for Computing Machinery

Place of publication

New York, NY, USA

Event location

Como, Italy

Event date

Aug 27, 2017

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—