Academic Style Marker Ontology Design

Viacheslav Lanin and Sofia Philipson

National Research University Higher School of Economics, Perm, Russian Federation

Keywords: Style Marker, Scientific Paper, Ontology, DSL.

Abstract: As any other genre, academic paper can be characterized by its own specific rules and fitches. The authors

assume that academic style fitches called in this research “style markers” can be modelled by means of

ontology engineering. The article is aimed at describing the academic style markers ontology design and its

practical using. The designed ontology is divided into two levels. The first level provides information about

linguistic terms and the second level consists of style markers, which were suggested by experts in linguistic.

It is assumed that two tasks will be solved on the basis of developed ontology. The first task is generating

lexical-semantic templates, which is used to identify the list of markers in a text. Due to ontology approach

and Domain Specific Language (DSL) technologies applying users can be able to extend and modify marker

templates. The second task is developing an expert system rules for text style enhancement.

1 INTRODUCTION

The contribution of research results in scientific

publications is the most significant academic

performance indicator of scholars and research co-

workers. Papers written in English extend the

audience match but non-native scholars usually face

some difficulties connected with the conventions of

formal academic writing style when write in English.

A huge number of researchers have investigated

academic writing features in general (Wallwork,

2016; Biber & Gray, 2016; Strongman, 2014;

Castelló & Donahue, 2012; Hyland, 2009), in terms

of grammar (Wallwork, 2013), structure (Sawaki,

2016), genre (Bruce, 2008) and other crucial features

of academic writing. Much of the previous research

projects were focused on identifying and evaluating

the features of academic writing implementing cross-

cultural and cross-linguistic approaches (Lakić et al.,

2015). However, not enough attention has been paid

to building a systematic approach to academic writing

as a segment of academic discourse both oral and

written as a functional style, a segment of the

language in which coherence and formality of the

discourse being its most prominent features

determine the choice of other elements.

Literature review has shown that

recommendations given in guides and handbooks for

both competent and novice academic writers in

English (Belcher, 2009; Bailey, 2011) are not

systematized and sometimes even have obvious

internal contradictions.

Typical webpages of university writing centers

define academic writing as having several distinctive

features: “Academic writing is to some extent:

complex, formal, objective, explicit, hedged, and

responsible. It uses language precisely and

accurately. It is also well organised and planned”.

However, such definitions do not help understand

importance of particular of academic writing features

and their hierarchy. Therefore, the purpose of our

research is developing the ontological system of

academic writing features. In this project (Lanin,

2015) we attempt to extract style markers, define

interrelations between them and design the style

model of academic English. Investigation of

hierarchical relations between style elements are also

crucial as it helps to determine their frequency

occurrence in English scientific texts and describe

usage pattern of these elements on the texts pieces of

different levels.

Traditionally, functional styles have been

explored by analyzing few texts representative for

particular style while at the moment more valid data

might be received by analysis of large corpora using

computer technologies. This enables the processing

of huge corpora and get empirical material which can

help create language patterns, study language

consistency, and describe linguistic phenomena

typical of a particular language area, i.e. derivation of

Lanin V. and Philipson S.

Academic Style Marker Ontology Design.

DOI: 10.5220/0006501701890194

In Proceedings of the 9th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (KEOD 2017), pages 189-194

ISBN: 978-989-758-272-1

style markers. In the paper, we adopt the term ‘style

marker’ instead of a broader ‘style feature’ due to

several reasons. 'Style feature’ represents a more

vague conception of a characteristic while ‘style

marker’ being ‘a sign’. Therefore, style markers in

this paper are considered as main features of

academic English in its linguistics meaning.

The data, collected from corpus annotated in

accordance with the style markers identified by

experts, give the information about the frequency of

occurrence of the elements and their leading or minor

role in building academic functional style.

2 RELATED WORK

One of the actively developing branches of theoretical

and applied stylistics is a complex analysis of English

scientific papers published in most leading peer-

reviewed academic journals conducted through

processing and comparative stylistics study and

critical discourse analysis of the large text corpora.

The analysis of English academic text produced by

non-native author offers the greatest challenge of

corpus linguistics research and the field of software

development for corpus analysis (Jeffries, 2009). It is

worth saying that there has been little quantitative

analysis of large corpora of written academic speech.

What is not yet clear is the hierarchy of linguistic

elements of academic writing what is a serious

obstacle to describing English of the particular

subject area with certainty, identify key features and

study usage pattern. The usage of computer

technologies simplifies statistical processing of large

corpora in linguistic research.

At the moment, a great variety of tools for corpus

processing exist. The most widespread of them are

AntConc, WordSmith Tools, Gate Developer, Sketch

Engine and CQPweb. There are specialized solutions

for academic papers style analysis, for example

project Fapas (Full Automatic Paper Analysis

System) (Scholz & Conrad, 2011).

It is also possible to find projects connected with

the creation of ontologies, which describe linguistic

domain (Zagorulko et al., 2010). One of them is

GOLD ontology which is General Ontology for

Linguistic Description (Farrar, 2003). It gives the

description of linguistic basis including most

foundational categories and relation between them.

The ontology is connected with SUMO (Standard

Upper Merged Ontology) (Pease et al. 2002) which is

based on four main domains: expressions, grammar,

data constructs, and metaconcepts.

Figure 1: Visualization fragment of ontology concepts.

The category expressions mean the physically

accessible aspects of language. The base for this

aspect was taken from SUMO and to the concept

LinguisticExpression were added new ones like

WrittenLinguisticExpressions and

SpokenLinguisticExpressions.

Grammar category includes the abstract

properties and relations of language, the domain that

is primary interest to linguists. It means that anything

expressed by a grammatical system is represented by

the concept GrammaticalCategory.

Data constructs are used by linguists to analyze

language data, such as paradigms, lexicons and

feature structures. Metaconcepts are the most basic

concepts of linguistic analysis, including language

itself. There are many ways in which language can

be viewed and without a working concept of the

language, an ontology cannot be used to describe and

compare data of different languages. Language was

defined as the set of data associated with a common

grammatical pattern. Generally speaking, the

ontology tries to describe all the aspect of the natural

language which can be applied to all languages.

Another example of ontology is also from

linguistic field but it is concentrated on computational

linguistics. Developed ontology is built on the basis

of scholarly knowledge ontology and is divided into

five hierarchies “whole-part” which are connected to

each other with associative relations. The subject of

investigation of computational linguistics is the

properties and the systems of linguistics units,

operations, connected with their functioning in the

process of communication, and application processes

replied to defined request.

3 THEORETICAL

BACKGROUND

The theoretical foundation of the system described in

this paper consists of a list of style markers that were

selected from reference and study materials, Internet

resources about academic writing as well as scientific

papers devoted to this topic. All markers from this list

can be divided into three main groups: lexical

markers, grammar markers, syntactic markers.

Lexical markers include three types of features:

• specific words and terminology (high frequency

of terminology; usage of abstract semantic verbs,

desemantisized verbs, intensifying adverbs; low

frequency of personal pronouns you, he, she);

• words corresponding to specific word-formation

constructions (nouns with -or suffix, commonly used

in terminology; abstract nouns derived by suffixes -

ment, -ness, -tion, etc.);

• words in specific part of speech (high frequency

of nouns, low frequency of pronouns).

Two types of features that belong to grammar

markers category are:

 wide usage of verbs in the Passive Voice;

 presumable prevalence of verbs in Present

Tense.

Syntactic markers can also be classified into two

types:

 features described by syntactic structures

(simple, complex or compound sentence

structure; prepositive and postpositive

attributes by most of the nouns; possible

prevalence of prepositive attributes in technical

texts);

 specific conjunctions, linking expressions, etc.

(subordinating and correlative conjunctions;

archaisms thereby, therewith, hereby;

prepositional phrases; means of logical

expressions).

Most of these features can be automatically

annotated using lexical-syntactic patterns, although

absolute accuracy cannot be guaranteed. That’s is

why expert control and means of manual annotation

correction are highly desirable for system

implementation. Flexibility of the system

components is also important for development and

further testing and debugging due to specificity of

academic style feature tagging and natural language

processing in general.

Currently our system annotates texts based on all

of the described style markers except terminology and

sentence structure. Although some components are

still being tested, recent resulting annotation sets

provide enough information to analyze academic

writing and deepen the studies about some of the

features.

For the present style markers are represented as

desperate data set. Language is dynamic and always

developing system, so besides markers

systematization the method should give the

opportunity of enlargement and adaptation.

4 ACADEMIC STYLE MARKER

ONTOLOGY

In the paper the way of regulation and systematization

of disparate data set called ontology is described. The

ontology reveals dependences between entities in the

form of style markers, and any existing

interconnections are indicated. Thus, a huge variety

of different style markers turn into a controlled

system which then can be used as a part of larger

project focused on improving the quality of text

annotating.

The combined approach to class hierarchy

development, which is based on dealing with the most

significant terms where the developer then

categorizes or typifies them, was chosen to design the

ontology.

After the data analysis of GOLD ontology was

done, it was decided to design the ontology which

will be based on the part of GOLD ontology,

describing lexical and syntax terms, i.e. the most

relevant terms of written English.

Figure 2: Main parts of ontology.

All aspects which were derived by experts earlier are

the part of subclasses called LinguisticProperty and

LinguisticUnit. It was decided to create the main

classes which are: LinguisticPropery and

LinguisticUnits – linguistic terms, Aspect – aspects

derived by experts and StyleMarker – markers which

were also given by experts. Class StyleMarker

represents particular style markers, which are the

features of written academic English. This class is

made of terms that were described earlier. Subclasses

of StyleMarker include individuals, which are style

markers of written academic English. Markers have

not any relations between each other but they are

connected with particular aspect, which them

expresses. Experts derive six main aspects of

academic English. They are not connected with each

other like markers, however they are connected with

terms of StlyleMarker subclasses as well as Linguistic

subclasses. At the current moment, there is only

hierarchic dependence between classes and

subclasses. Individuals of ontology are particular

examples of style markers of written academic

English.

It was decided to link the terms of classes named

Aspect and StyleMarker with properties express and

it’s inverse property isExpressedBy. Every class of

aspects is expressed by subclass/subclasses of

particular style marker of written academic English.

Now we will talk about properties which are used for

linking of subclasses of Aspect with subclasses of

StyleMarker and LinguisticPropetry. For example,

the graph, which is represented on the figure 4, shows

the relations of aspect named Adverb.

Figure 3: Ontology concepts hierarchy.

Figure 4: Relations and instances of «Adverb» aspect.

It can be seen that this class is a subclass of class

named Aspect and class Adverb is connected with a

subclass IntensifyingAdverb of class StyleMarker.

The following properties were created: Adverb

isExpressedBy IntensifyingAdverb, which means that

aspect Adverb is expressed by style marker

Intensifying adverb. Besides there is one more

property hasIndividual and particular style marker’s

individuals. If we talk about connection between

linguistic terms and aspects there are properties called

include and isPartOf: Adverb isPartOf Adverbial and

it’s inverse sentence with property include.

GOLD

Aspects

SUMO

Stylemarkers

Class Verb is subclass of class named Aspect, it is

connected with subclasses Voice, Tense,

DesimanticisedVerb of class StyleMarker. The

properties are: Verb isExpressedBy Voice, Tense,

DesimanticisedVerb and Voice, Tense,

DesimanticisedVerb express Verb. The relations

between linguistics and aspect are following: Verb

isPartOf Verbal and Verbal include Verb.

Graph of aspect called Cohesiveness shows

relations between style markers and linguistic terms.

This aspect is expressed by subclasses

ComplexСonjunction, ComplexPreposition,

LogicConnector and Archaism. This aspect is part of

Functor linguistic term.

Aspect Nominalization is expressed by Noun and

NounWithAbstractSuffix. NounWithAbstractSuffix

has the property hasSuffix, which refers to abstract

suffixes, which are represented as individuals of class

AbstractSuffix. Aspect Nominalization is a part of

linguistic term Noun.

Aspect PersonalStance is expressed by style

markers PersonalNoun and DemonstrativePronoun.

Aspect is connected with linguistic term

PersonalPronoun. Style markers has property

hasIndividual, that is shown on the figure 5.

Figure 5: Relations and instances of PersonalStance aspect.

5 PATTERN GENERATION

BASED ON

DSL-TECHNOLOGIES

Ontology is necessary not only for style markers

systematization but also as the foundation of lexical-

semantic patterns (IJntema, W. et al., 2012)

generation. Currently, JAPE-templates (Java

Annotation Patterns Engine) of GATE (Cunningham

et al., 2011) text processing system are used. Rule

generation architecture is demonstrated on figure 6.

Figure 6: The architecture of the lexical-semantic pattern

generator.

Protégé ontology editor is used for ontology

describing and its representation in the OWL format.

Validator is the component which is meant for

accuracy check of user’s models. While designing a

model, the user can make some mistakes or make

models which are not satisfy the ontology limits

constraints. Generator is the component responsible

for code generation on target language. Generator is

used for transformation of user’s models into textual

representation on the description language of lexical-

semantic patterns as well as file generation into the

formats of the computer linguistic systems for

example JAPE. To extend the interoperability the

system gives users the opportunity of manual

determining the transformation rules. It is crucial on

this level of metamodel to make text pattern for every

language element in accordance to which code

generation would be implemented. Text pattern

includes the statistic part which is not depended on

certain model and the dynamic part, which makes

possible the reference to attributes values of different

DSL-constructions.

General algorithm of search pattern generating on

the basis of ontology consists of several steps:

1. Get all the individuals of the subclasses

which main class is StyleMarker.

Protégé ontology editor

Validator

Generator

Generation rules

2. Get data property Template, which contains

JAPE - expression to search particular

marker.

3. Get JAPE –expression or name of file which

is aimed on searching of marker.

4. Combine expressions with the help of OR-

operator in one total JAPE - expression, get

to file name, consisted style marker search

implementation.

To perform these steps SPARQL-queries should be

written.

6 CONCLUSIONS

The standard tools and software applications are used

to design the ontology which simplifies the process of

development and decision maintenance process. The

described approach has an expanding property, i.e. in

order to add a new marker the user needs to add its

description and the identification rule will be

generated automatically. Moreover, using the

linguistics level, described in the ontology, makes the

description of related domains possible.

ACKNOWLEDGEMENTS

The article was prepared within the framework of the

Academic Fund Program at the National Research

University Higher School of Economics (HSE) in

2017 (grant № 17-05-0020) and by the Russian

Academic Excellence Project "5-100".

REFERENCES

Bailey, S., 2011. Academic Writing: A Handbook for

International Students, Taylor & Francis.

Belcher, W.L., 2009. Writing Your Journal Article in

Twelve Weeks: A Guide to Academic Publishing

Success, SAGE Publications.

Biber, D. & Gray, B., 2016. Grammatical Complexity in

Academic English: Linguistic Change in Writing,

Cambridge University Press.

Bruce, I., 2008. Academic Writing and Genre: A Systematic

Analysis, Bloomsbury Publishing.

Castelló, M. & Donahue, C., 2012. University Writing:

Selves and Texts in Academic Societies, Emerald.

Cunningham, H., Maynard, D. & Bontcheva, K., 2011. Text

Processing with GATE (Version 6), University of

Sheffield.

Farrar, S. & Langendoen, T., 2003. A linguistic ontology

for the semantic web. Glot International, 7, pp.97–100.

Hartley, J., 2008. Academic writing and publishing: A

practical handbook, Routledge.

Hyland, K., 2009. Academic Discourse: English In A

Global Context, Bloomsbury Publishing.

IJntema, W. et al., 2012. A lexico-semantic pattern

language for learning ontology instances from text.

Web Semantics: Science, Services and Agents on the

World Wide Web, 15, pp.37–50.

Jeffries, L., 2009. Critical Stylistics: The Power of English,

Palgrave Macmillan.

Lakić, I., Živković, B. & Vuković, M., 2015. Academic

Discourse across Cultures, Cambridge Scholars

Publishing.

Lanin, V., Strinyuk, S. & Shuchalova, Y., 2015. Academic

papers evaluation software. In 2015 9th International

Conference on Application of Information and

Communication Technologies (AICT). pp. 506–510.

Pease, A., Niles, I. & Li, J., 2002. The Suggested Upper

Merged Ontology: A Large Ontology for the Semantic

Web and its Applications. Imagine, 28, pp.7–10.

Sawaki, T., 2016. Analysing Structure in Academic Writing,

Palgrave Macmillan UK.

Scholz, T. & Conrad, S., 2011. Style Analysis of Academic

Writing, pp.246–249.

Strongman, L., 2014. Academic Writing, Cambridge

Scholars Publisher.

Wallwork, A., 2013. English for Academic Research:

Grammar, Usage and Style, Springer, Boston, MA.

Wallwork, A., 2016. English for Writing Research Papers,

Springer, Cham.

Zagorulko, Y., Borovikova, O. & Zagorulko, G., 2010.

Knowledge Portal on Computational Linguistics :

Content-Based Multilingual Access to Linguistic

Information Resources. In Selected topics in Applied

Computer Science. Proceedings of the 10th WSEAS

International Conference on Applied Computer Science

(ACS’10). pp. 255–262.