Machine-based mapping of innovation portfolios

Research output: Contribution to conferencePaperAcademicpeer-review

Abstract

Machine learning techniques show a great promise for improving innovation portfolio management. In this paper we experiment with different methods to classify innovation projects of a high-tech firm as either explorative or exploitative, and compare the results with a manual, theory-based mapping of these projects and with expert classification. We find that by combining a high-information extraction method with a decision tree or maximum entropy algorithm, higher levels of accuracy can be reached. Opportunities and limitations of different methods are discussed.
Original languageEnglish
Pages667-671
Number of pages4
Publication statusPublished - 10 Sep 2017
Event18th International CINet Conference - Potsdam, Germany
Duration: 10 Sep 201712 Sep 2017
Conference number: 18
http://www.continuous-innovation.net/events/conferences/2017.html#0

Conference

Conference18th International CINet Conference
CountryGermany
CityPotsdam
Period10/09/1712/09/17
Internet address

Fingerprint

Innovation
Portfolio management
High-tech firms
Maximum entropy
Experiment
Information extraction
Machine learning
Decision tree

Cite this

de Visser, M., Miao, S., Englebienne, G., Sools, A. M., & Visscher, K. (2017). Machine-based mapping of innovation portfolios. 667-671. Paper presented at 18th International CINet Conference, Potsdam, Germany.
de Visser, Matthias ; Miao, Shengfa ; Englebienne, Gwenn ; Sools, Anna Maria ; Visscher, Klaasjan . / Machine-based mapping of innovation portfolios. Paper presented at 18th International CINet Conference, Potsdam, Germany.4 p.
@conference{b1d6559dc0a3483899a94e8fcc306d32,
title = "Machine-based mapping of innovation portfolios",
abstract = "Machine learning techniques show a great promise for improving innovation portfolio management. In this paper we experiment with different methods to classify innovation projects of a high-tech firm as either explorative or exploitative, and compare the results with a manual, theory-based mapping of these projects and with expert classification. We find that by combining a high-information extraction method with a decision tree or maximum entropy algorithm, higher levels of accuracy can be reached. Opportunities and limitations of different methods are discussed.",
author = "{de Visser}, Matthias and Shengfa Miao and Gwenn Englebienne and Sools, {Anna Maria} and Klaasjan Visscher",
year = "2017",
month = "9",
day = "10",
language = "English",
pages = "667--671",
note = "18th International CINet Conference ; Conference date: 10-09-2017 Through 12-09-2017",
url = "http://www.continuous-innovation.net/events/conferences/2017.html#0",

}

de Visser, M, Miao, S, Englebienne, G, Sools, AM & Visscher, K 2017, 'Machine-based mapping of innovation portfolios' Paper presented at 18th International CINet Conference, Potsdam, Germany, 10/09/17 - 12/09/17, pp. 667-671.

Machine-based mapping of innovation portfolios. / de Visser, Matthias ; Miao, Shengfa ; Englebienne, Gwenn ; Sools, Anna Maria; Visscher, Klaasjan .

2017. 667-671 Paper presented at 18th International CINet Conference, Potsdam, Germany.

Research output: Contribution to conferencePaperAcademicpeer-review

TY - CONF

T1 - Machine-based mapping of innovation portfolios

AU - de Visser, Matthias

AU - Miao, Shengfa

AU - Englebienne, Gwenn

AU - Sools, Anna Maria

AU - Visscher, Klaasjan

PY - 2017/9/10

Y1 - 2017/9/10

N2 - Machine learning techniques show a great promise for improving innovation portfolio management. In this paper we experiment with different methods to classify innovation projects of a high-tech firm as either explorative or exploitative, and compare the results with a manual, theory-based mapping of these projects and with expert classification. We find that by combining a high-information extraction method with a decision tree or maximum entropy algorithm, higher levels of accuracy can be reached. Opportunities and limitations of different methods are discussed.

AB - Machine learning techniques show a great promise for improving innovation portfolio management. In this paper we experiment with different methods to classify innovation projects of a high-tech firm as either explorative or exploitative, and compare the results with a manual, theory-based mapping of these projects and with expert classification. We find that by combining a high-information extraction method with a decision tree or maximum entropy algorithm, higher levels of accuracy can be reached. Opportunities and limitations of different methods are discussed.

M3 - Paper

SP - 667

EP - 671

ER -

de Visser M, Miao S, Englebienne G, Sools AM, Visscher K. Machine-based mapping of innovation portfolios. 2017. Paper presented at 18th International CINet Conference, Potsdam, Germany.