---
_id: '44323'
abstract:
- lang: eng
text: "Reading between the lines has so far been reserved for humans. The present
dissertation addresses this research gap using machine learning methods.\r\nImplicit
expressions are not comprehensible by computers and cannot be localized in the
text. However, many texts arise on interpersonal topics that, unlike commercial
evaluation texts, often imply information only by means of longer phrases. Examples
are the kindness and the attentiveness of a doctor, which are only paraphrased
(“he didn’t even look me in the eye”). The analysis of such data, especially the
identification and localization of implicit statements, is a research gap (1).
This work uses so-called Aspect-based Sentiment Analysis as a method for this
purpose. It remains open how the aspect categories to be extracted can be discovered
and thematically delineated based on the data (2). Furthermore, it is not yet
explored how a collection of tools should look like, with which implicit phrases
can be identified and thus made explicit\r\n(3). Last, it is an open question
how to correlate the identified phrases from the text data with other data, including
the investigation of the relationship between quantitative scores (e.g., school
grades) and the thematically related text (4). Based on these research gaps, the
research question is posed as follows: Using text mining methods, how can implicit
rating content be properly interpreted and thus made explicit before it is automatically
categorized and quantified?\r\nThe uniqueness of this dissertation is based on
the automated recognition of implicit linguistic statements alongside explicit
statements. These are identified in unstructured text data so that features expressed
only in the text can later be compared across data sources, even though they were
not included in rating categories such as stars or school grades. German-language
physician ratings from websites in three countries serve as the sample domain.
The solution approach consists of data creation, a pipeline for text processing
and analyses based on this. In the data creation, aspect classes are identified
and delineated across platforms and marked in text data. This results in six datasets
with over 70,000 annotated sentences and detailed guidelines. The models that
were created based on the training data extract and categorize the aspects. In
addition, the sentiment polarity and the evaluation weight, i. e., the importance
of each phrase, are determined. The models, which are combined in a pipeline,
are used in a prototype in the form of a web application. The analyses built on
the pipeline quantify the rating contents by linking the obtained information
with further data, thus allowing new insights.\r\nAs a result, a toolbox is provided
to identify quantifiable rating content and categories using text mining for a
sample domain. This is used to evaluate the approach, which in principle can also
be adapted to any other domain."
author:
- first_name: Joschka
full_name: Kersting, Joschka
id: '58701'
last_name: Kersting
citation:
ama: Kersting J. Identifizierung quantifizierbarer Bewertungsinhalte und -kategorien
mittels Text Mining. Universität der Bundeswehr München ; 2023.
apa: Kersting, J. (2023). Identifizierung quantifizierbarer Bewertungsinhalte
und -kategorien mittels Text Mining. Universität der Bundeswehr München .
bibtex: '@book{Kersting_2023, place={Neubiberg}, title={Identifizierung quantifizierbarer
Bewertungsinhalte und -kategorien mittels Text Mining}, publisher={Universität
der Bundeswehr München }, author={Kersting, Joschka}, year={2023} }'
chicago: 'Kersting, Joschka. Identifizierung quantifizierbarer Bewertungsinhalte
und -kategorien mittels Text Mining. Neubiberg: Universität der Bundeswehr
München , 2023.'
ieee: 'J. Kersting, Identifizierung quantifizierbarer Bewertungsinhalte und -kategorien
mittels Text Mining. Neubiberg: Universität der Bundeswehr München , 2023.'
mla: Kersting, Joschka. Identifizierung quantifizierbarer Bewertungsinhalte und
-kategorien mittels Text Mining. Universität der Bundeswehr München , 2023.
short: J. Kersting, Identifizierung quantifizierbarer Bewertungsinhalte und -kategorien
mittels Text Mining, Universität der Bundeswehr München , Neubiberg, 2023.
date_created: 2023-05-02T12:54:00Z
date_updated: 2023-07-03T12:29:50Z
department:
- _id: '579'
- _id: '7'
language:
- iso: ger
page: '208'
place: Neubiberg
project:
- _id: '1'
grant_number: '160364472'
name: 'SFB 901: SFB 901'
- _id: '3'
name: 'SFB 901 - B: SFB 901 - Project Area B'
- _id: '9'
grant_number: '160364472'
name: 'SFB 901 - B1: SFB 901 - Subproject B1'
publication_status: published
publisher: 'Universität der Bundeswehr München '
related_material:
link:
- relation: supplementary_material
url: https://athene-forschung.unibw.de/145003
status: public
supervisor:
- first_name: Michaela
full_name: Geierhos, Michaela
id: '42496'
last_name: Geierhos
orcid: 0000-0002-8180-5606
title: Identifizierung quantifizierbarer Bewertungsinhalte und -kategorien mittels
Text Mining
type: dissertation
user_id: '58701'
year: '2023'
...
---
_id: '13435'
author:
- first_name: Edwin
full_name: Friesen, Edwin
last_name: Friesen
citation:
ama: 'Friesen E. Requirements Engineering im OTF-Computing: Informationsextraktion
und Unvollständigkeitskompensation mittels domänenspezifischer Wissensbasis.
Universität Paderborn; 2019.'
apa: 'Friesen, E. (2019). Requirements Engineering im OTF-Computing: Informationsextraktion
und Unvollständigkeitskompensation mittels domänenspezifischer Wissensbasis.
Universität Paderborn.'
bibtex: '@book{Friesen_2019, title={Requirements Engineering im OTF-Computing: Informationsextraktion
und Unvollständigkeitskompensation mittels domänenspezifischer Wissensbasis},
publisher={Universität Paderborn}, author={Friesen, Edwin}, year={2019} }'
chicago: 'Friesen, Edwin. Requirements Engineering im OTF-Computing: Informationsextraktion
und Unvollständigkeitskompensation mittels domänenspezifischer Wissensbasis.
Universität Paderborn, 2019.'
ieee: 'E. Friesen, Requirements Engineering im OTF-Computing: Informationsextraktion
und Unvollständigkeitskompensation mittels domänenspezifischer Wissensbasis.
Universität Paderborn, 2019.'
mla: 'Friesen, Edwin. Requirements Engineering im OTF-Computing: Informationsextraktion
und Unvollständigkeitskompensation mittels domänenspezifischer Wissensbasis.
Universität Paderborn, 2019.'
short: 'E. Friesen, Requirements Engineering im OTF-Computing: Informationsextraktion
und Unvollständigkeitskompensation mittels domänenspezifischer Wissensbasis, Universität
Paderborn, 2019.'
date_created: 2019-09-20T14:58:49Z
date_updated: 2022-01-06T06:51:36Z
department:
- _id: '36'
- _id: '1'
- _id: '579'
language:
- iso: ger
project:
- _id: '1'
name: SFB 901
- _id: '3'
name: SFB 901 - Project Area B
- _id: '9'
name: SFB 901 - Subproject B1
publisher: Universität Paderborn
status: public
supervisor:
- first_name: Eyke
full_name: Hüllermeier, Eyke
id: '48129'
last_name: Hüllermeier
- first_name: Michaela
full_name: Geierhos, Michaela
id: '42496'
last_name: Geierhos
orcid: 0000-0002-8180-5606
title: 'Requirements Engineering im OTF-Computing: Informationsextraktion und Unvollständigkeitskompensation
mittels domänenspezifischer Wissensbasis'
type: bachelorsthesis
user_id: '477'
year: '2019'
...
---
_id: '89'
abstract:
- lang: eng
text: The vision of OTF Computing is to have the software needs of end users in
the future covered by an automatic composition of existing software services.
Here we focus on natural language software requirements that end users formulate
and submit to OTF providers as requirement specifications. These requirements
serve as the sole foundation for the composition of software; but they can be
inaccurate and incomplete. Up to now, software developers have identified and
corrected these deficits by using a bidirectional consolidation process. However,
this type of quality assurance is no longer included in OTF Computing - the classic
consolidation process is dropped. This is where this work picks up, dealing with
the inaccuracies of freely formulated software design requirements. To do this,
we developed the CORDULA (Compensation of Requirements Descriptions Using Linguistic
Analysis) system that recognizes and compensates for language deficiencies (e.g.,
ambiguity, vagueness and incompleteness) in requirements written by inexperienced
end users. CORDULA supports the search for suitable software services that can
be combined in a composition by transferring requirement specifications into canonical
core functionalities. This dissertation provides the first-ever method for holistically
recording and improving language deficiencies in user-generated requirement specifications
by dealing with ambiguity, incompleteness and vagueness in parallel and in sequence.
author:
- first_name: Frederik Simon
full_name: Bäumer, Frederik Simon
id: '38837'
last_name: Bäumer
citation:
ama: Bäumer FS. Indikatorbasierte Erkennung und Kompensation von ungenauen und
unvollständig beschriebenen Softwareanforderungen. Universität Paderborn;
2017. doi:10.17619/UNIPB/1-157
apa: Bäumer, F. S. (2017). Indikatorbasierte Erkennung und Kompensation von ungenauen
und unvollständig beschriebenen Softwareanforderungen. Universität Paderborn.
https://doi.org/10.17619/UNIPB/1-157
bibtex: '@book{Bäumer_2017, title={Indikatorbasierte Erkennung und Kompensation
von ungenauen und unvollständig beschriebenen Softwareanforderungen}, DOI={10.17619/UNIPB/1-157}, publisher={Universität
Paderborn}, author={Bäumer, Frederik Simon}, year={2017} }'
chicago: Bäumer, Frederik Simon. Indikatorbasierte Erkennung und Kompensation
von ungenauen und unvollständig beschriebenen Softwareanforderungen. Universität
Paderborn, 2017. https://doi.org/10.17619/UNIPB/1-157.
ieee: F. S. Bäumer, Indikatorbasierte Erkennung und Kompensation von ungenauen
und unvollständig beschriebenen Softwareanforderungen. Universität Paderborn,
2017.
mla: Bäumer, Frederik Simon. Indikatorbasierte Erkennung und Kompensation von
ungenauen und unvollständig beschriebenen Softwareanforderungen. Universität
Paderborn, 2017, doi:10.17619/UNIPB/1-157.
short: F.S. Bäumer, Indikatorbasierte Erkennung und Kompensation von ungenauen und
unvollständig beschriebenen Softwareanforderungen, Universität Paderborn, 2017.
date_created: 2017-10-17T12:41:08Z
date_updated: 2022-01-06T07:04:05Z
department:
- _id: '579'
- _id: '36'
- _id: '1'
doi: 10.17619/UNIPB/1-157
language:
- iso: ger
project:
- _id: '1'
name: SFB 901
- _id: '9'
name: SFB 901 - Subprojekt B1
- _id: '3'
name: SFB 901 - Project Area B
publisher: Universität Paderborn
status: public
supervisor:
- first_name: Michaela
full_name: Geierhos, Michaela
id: '42496'
last_name: Geierhos
orcid: 0000-0002-8180-5606
title: Indikatorbasierte Erkennung und Kompensation von ungenauen und unvollständig
beschriebenen Softwareanforderungen
type: dissertation
user_id: '14931'
year: '2017'
...