Arguments in Parliamentary Negotiation: A Study of Verbatim

Records

Mare Koit

Institute of Computer Science, University of Tartu, Narva mnt 18, Tartu, Estonia

Keywords: Argument, Negotiation, Parliament, Verbatim Record, Knowledge Representation, Annotated Corpus.

Abstract: Verbatim records of sittings of the Estonian Parliament are studied in this paper. The general structure of the

discussions is presented. Arguments used in negotiation are considered as consisting of premises and claims.

The relations between the arguments (attack, rebuttal, support) are determined with the aim to create a corpus

where arguments are annotated. Our further aim is the automatic recognition of arguments and their relations

in Estonian political texts. To our knowledge, this is the first attempt towards modelling Estonian political

argumentation.

1 INTRODUCTION

Parliamentary discourse is an important resource

because it contains impactful information and special,

formalised and often persuasive and emotional

language. The empirical study of parliamentary

discourse contributes to an understanding of how

policy issues are framed. The study can also be related

to comparative assessments of the deliberative

performance of different parliaments (Bara et al. 2007).

There are many ongoing initiatives for compiling

digital collections of parliament data (Working 2017).

The recent CLARIN-PLUS survey on parliament data

has identified over 20 corpora of parliamentary

records, with over half of them being available within

the CLARIN infrastructure (Fišer, Lenardič 2018).

The data can be used for linguistic, historical,

political, sociological etc. research.

Parliamentary debates are full of arguments.

Analysing argumentation from a computational

linguistics point of view has recently led to a new ﬁeld

called argumentation mining. The review of Atkinson

et al. (2015) considers the development of artiﬁcial

tools that capture the human ability to argue. Such

systems can be used when modelling political

argumentation since they are able to extract

arguments and relations between them automatically.

In the current paper, we study negotiations on

motions in the Estonian Parliament (Riigikogu) based

https://orcid.org/0000-0002-7318-087X

on verbatim records of the sittings (in Estonian). In

the records, repetitions and disﬂuencies are omitted,

while supplementary information such as speaker

names are added. In the paper, we are looking for

arguments presented in negotiations for and against a

motion. To our knowledge, this is the first attempt to

analyse and model the formal structure and relations

of arguments in Estonian political discourse.

The paper is structured as follows. Section 2

describes related work. In Section 3, we examine one

randomly selected discussion in Riigikogu by using

verbatim records of the sittings. We consider the

arguments presented by the members of the parliament

and determine the inter-argument relations. Section 4

discusses some problems of annotating the arguments

with the aim to create a corpus in order to prepare

automatic recognition of arguments. Section 5 draws

conclusions and figures out the future work.

2 RELATED WORK

A lot of work has been done when studying political

discourse.

Bara et al. (2007) examine one UK parliamentary

debate on abortion with the aim to identify the

prominent themes in debate and to assess how far

speakers who favour different positions adopt a

distinct pattern of discourse.

822

Koit, M.

Arguments in Parliamentary Negotiation: A Study of Verbatim Records.

DOI: 10.5220/0009128908220828

In Proceedings of the 12th International Conference on Agents and Artiﬁcial Intelligence (ICAART 2020) - Volume 2, pages 822-828

ISBN: 978-989-758-395-7; ISSN: 2184-433X

Vilares and He (2017) propose a Bayesian

modelling approach where topics are modelled as

latent variables. The model is evaluated on debates

from the House of Commons of the UK Parliament.

This is the ﬁrst novel work towards topic modelling.

Addawood et al. (2017) investigate the question

of whether opinion mining techniques can be used on

Congressional debates or not.

Venkata et al. (2018) provide a dataset for the

synopsis of Indian parliamentary debates and perform

stance classification of speeches, identifying if the

speaker is supporting the bill. Based on manual

analysis of the debates, they develop an annotation

scheme of four mutually exclusive categories to

analyse the purpose of the speeches.

Special attention has been paid to argumentation

in political discussions.

Walker et al. (2012) analyse deliberations and

debates by using the Internet Argument Corpus. The

corpus includes the posts from a website for political

debate where the debates are annotated for

argumentative markers like degrees of agreement

with previous post, cordiality, audience direction,

combativeness, assertiveness, emotionality of

argumentation, and sarcasm.

In parliamentary discourse, politicians expound

their beliefs and ideas through argumentation, and to

persuade the audience, they highlight some aspect of

an issue, which is commonly known as framing.

Naderi and Hirst (2015) examine how to identify

framing strategies in argumentative political speech.

They use a corpus of speeches from the Canadian

Parliament, and examine the statements with respect

to the position of the speaker towards the discussed

topic (Pro, Con, or No stance).

Petukhova et al. (2015) use the Information State

Update (ISU) approach to model the arguments of the

debaters and the support/attack links between them as

part of the formal representations of a participant’s

information state. They consider the identiﬁcation of

claims and evidence relations to their premises as an

argument mining task. The ISU model provides

procedures for incorporating beliefs and expectations

shared between speaker and hearers in the tracking

model.

Lippi and Torroni (2016a) investigate how to

improve claim detection for argument mining, by

employing features from text and speech in

combination. They develop a machine learning

classiﬁer and train it on an original dataset based on

the 2015 UK political elections debate.

Petukhova et al. (2017) have collected the

Metalogue Debate Corpus that includes 400

arguments from six different bilingual

(English/Greek) speakers. The corpus is used to

design a Virtual Debate Coach, in order to train young

parliamentarians on how to debate successfully.

Although it is often difﬁcult to deﬁne clear properties

of persuasive debate, there are certain linguistic,

prosodic and body language features that correlate

with human judgments of such behaviour.

Haddadan et al. (2018) present annotation

guidelines for annotating arguments (their premises

and claims) in political debates. The dataset is taken

from the Commission on Presidential Debates

website which is an independent non-proﬁt

corporation sponsoring U.S. presidential and vice-

presidential debates.

Menini et al. (2018) apply argumentation mining

techniques, in particular relation prediction, to study

political speeches – monologues, where there is no

direct interaction between opponents. They have

created a corpus, based on the transcription of

speeches and ofﬁcial declarations issued by Nixon

and Kennedy during 1960 Presidential campaign, of

argument pairs annotated with the support and attack

relations. They use a tool called OVA+ (Janier et al.

2014), an on-line interface for the manual analysis of

natural language arguments.

Many other studies have contributed to

development of formalisms and tools for analysing

arguments.

Chesňevar et al. (2006) introduce a speciﬁcation

for an argument interchange format intended for

representation and exchange of data between various

argumentation tools and agent-based applications.

Reed et al. (2008) describe a written corpus of

argumentative reasoning. Arguments have been

analysed using techniques from argumentation theory

and have been marked up. The authors present

experiences with initial pilot data collection, which

raised a number of key questions that frame

challenges for argument corpora in general.

Besnard and Hunter (2014) consider a deductive

argument as a pair where the ﬁrst item is a set of

premises, the second item is a claim, and the premises

entail the claim. This can be formalised by assuming

a logical language for the premises and the claim, and

logical entailment (or consequence relation) for

showing that the claim follows from the premises.

Stab and Gurevych (2014) present a novel

approach to model arguments, their components and

relations in persuasive essays in English. The

annotation scheme includes the annotation of claims

and premises as well as support and attack relations

for capturing the structure of argumentative

Arguments in Parliamentary Negotiation: A Study of Verbatim Records

823

discourse. The authors conduct a manual annotation

study with three annotators on 90 persuasive essays.

Amgoud et al. (2015) consider an argument as

reasons in favour of a claim. It is made of three parts:

a set of premises representing the reasons, a

conclusion representing the supported claim, and a

connection showing how the premises lead to the

conclusion. They propose a language for representing

such arguments that captures the various forms of

arguments encountered in natural language, and

demonstrate that it is possible to represent attack and

support relations between arguments as formulas of

the same language.

MARGOT (Mining ARGuments frOm Text) is a

web server for the automatic extraction of arguments

from text (Lippi, Torroni 2016b). It focuses on

detection of argument components (claim and

evidence). A claim is a general, typically concise

statement that directly supports or contests a topic

under debate, whereas evidence is a text segment that

directly supports a claim. The tool currently supports

only English.

Atkinson et al. (2017) summarise that recent

developments are leading to technology for artiﬁcial

argumentation, in the legal, medical, and e-

government domains, and interesting tools for

argument mining, for debating technologies, and for

argumentation solvers are emerging. The extracted

arguments will represent the nodes in an argument

graph and the links are the relations between the

arguments.

3 ARGUMENTS PRESENTED BY

THE MEMBERS OF

RIIGIKOGU

In this paper, we are analysing the verbatim records

of discussions held in the Parliament of Estonia, in

order to figure out the structure of the arguments

presented in negotiation. Our current aim is to create

a corpus where arguments are annotated. Such a

corpus will contribute to the automatic recognition of

arguments and can thereby promote studies on

Estonian political discourse.

3.1 Empirical Material

Our empirical material is formed by the records of the

Parliament of Estonia – Riigikogu (cf. Riigikogu). An

important task of the Riigikogu is the passing of acts

and resolutions. Acts are the result of work in

multiple stages. The first stage of legislation involves

the drafting of a bill (a draft act). During the second

stage, the bill is initiated in the Riigikogu. The bill

will then pass three readings (in exceptional cases

two), during which it is refined and amended. The

proceeding of a bill is managed by the relevant

leading committee. After having been passed by the

Riigikogu, the act is sent to the President of the

Republic for proclamation, and is then published in

State Gazette. The general structure of the process in

Riigikogu is presented in Fig. 1 (cf. Koit et al. 2019).

The authors of turns are given in italics.

- - 1

reading

- - initiator – Government

Presenter (Minister): Report

{

MP: Question

Presenter: Giving information

}

Co-presenter (a member of leading committee): Report

{

MP: Question

Co-presenter: Giving information

}

- - negotiation

{

MP: argument

}

- - 2

reading

Presenter (a member of leading committee): Report

on amendment motions

{

MP: Question

Presenter: Giving information

}

- - negotiation

{

MP: argument

}

- - voting on amendment motions

- - 3

reading

- - negotiation

{

MP: argument

}

- - final voting

Figure 1: The general structure of discussions in Riigikogu.

The winding brackets ‘{‘ and ‘}’ connect a part that can be

repeated; ‘- -’ starts a comment; MP– any member of

Riigikogu. The authors of turns are given in italics.

Verbatim records of sittings of the Riigikogu (in

Estonian) are accessible on the web as pdf files. For

the current study, we have randomly chosen a draft

act on social care that was proceeded in 2018 (the

records of three sittings include 14,662 running words

in total). The aim of the act is to create an additional

supporting system for youth security in order to

decrease the rate of the unemployment of young

ICAART 2020 - 12th International Conference on Agents and Artiﬁcial Intelligence

824

people (16–26 years) which is twice the average

unemployment (14.1% vs 6%). The debate ends after

voting with adopting of the act by the members of

Riigikogu (MPs).

3.2 Arguments

An argument is a series of statements (in a natural

language), called the premises, intended to determine

the degree of truth of another statement – the claim.

Therefore, an argument consists of two parts: of one

or more premises and of a claim. These parts can be

presented in one or more sentences. There are three

types of relations between the arguments: attack,

support, and rebuttal (Amgoud et al. 2015).

When analysing persuading essays, Stab and

Gurevych (2014) make a distinction between the

major claim and a claim. In parliamentary

discussions, we similarly can differentiate the major

claim and a claim of an arbitrary argument. The major

claim together with its premises is given in the report

of Minister and it is always ‘to accept the bill’ (Fig.1).

As a rule, the claim of a supporting argument

presented in following negotiation, coincides with the

major claim. The claim of a rebutting argument is

opposite: do not accept the bill. The claim of an

attacking argument depends on a previous argument

that is under attack. The arguments and relations will

be illustrated by the following examples.

When taking the floor in negotiation, the members

of Riigikogu are always presenting their arguments in

more than one sentence. A premise and a claim are

located in different sentences. In some cases, the

claim is missing (is default), mostly when it coincides

with the major claim or, on the contrary, when it is

opposite. Typically, the arguments have more than

one premise.

In the analysed negotiations, premises and claims

of arguments were manually annotated by the author

of the paper, following (Stab, Gurevich 2014) and

(Amgoud et al. 2015). (In consequence, the

annotation is rather subjective.) Let us give some

examples of arguments and relations between them.

Major claim (‘to accept the bill’) and its premises

are given at the beginning of the first reading, in the

reports of Minister and the member of a leading

committee (Fig. 1). Example 1 presents some of the

premises given by Minister.

Example 1. Three premises of the major claim

presented by Minister.

(1)

Teatavasti on Eestis noorte töötuse määr ligi kaks korda

suurem kui keskmine töötuse määr ning viimasel ajal on see

vaatamata üldisele heale tööturu seisule hoopis suurenema

hakanud.

As known, the unemployment of young people is

twice the average unemployment and it is increasing

although the general situation on the work market is

good.

</premise>

(2)

Noortegarantii tugisüsteemi eesmärk ongi vähendada

mittetöötavate ja mitteõppivate noorte arvu.

The aim of the supporting system is to decrease

the number of the young people who are neither

working nor learning.

</premise>

(3)

Kuna tegemist on andmetöötlusega, siis on selle

süsteemi kasutamiseks vaja seaduslikku alust.

The legislative basis is needed because there is

data processing.

</premise>

The following counterarguments presented by

MPs in negotiations include different claims:

negation of the major claim (Example 2) and

derivatives of the major claim (Examples 3 and 4

where premises of the major claim are attacked).

Example 2. Argument rebutting the major claim.

<argument

- - rebutting

Eelnõu kohaselt võimaldatakse omavalitsustel

proaktiivselt pakkuda potentsiaalselt abi vajavatele noortele

inimestele tuge kas tööelu alustamiseks või haridusellu

naasmiseks.

The purpose of the bill is to provide proactive help

by a local government to young people who need

support for starting to work or for returning to

education.

</premise>

/---/

Kuid kes oleks osanud arvata, et selle hea eesmärgi

saavutamiseks minnakse nii kaugele, et hakatakse isikuandmeid

töötlema liiga massiliselt ehk hakatakse tegelema omaalgatusliku

nuhkimisega.

But who could believe that in order to achieve this

nice goal, personal data will be processed so

massively, i.e. people will be tracked down.

</premise>

Arguments in Parliamentary Negotiation: A Study of Verbatim Records

825

<claim>

Sellist seadust meile tegelikult vaja ei ole.

We don’t need such a law.

Kehtivas hoolekandeseaduses on olemas kõik hoovad

statistiliste andmete kogumiseks ja analüüsimiseks ning noorte

inimeste abistamiseks.

The valid law already includes all the instruments

necessary for collecting and analysing statistical

data.

</claim>

</argument

The argument in Example 2 strongly rebuts the

major claim (s. Amgoud et al. 2015).

Example 3. Argument attacking a reason of the

major claim.

<argument

- - attacking

Meie ees olev eelnõu on kõige ilmekam näide õigusnormide

loomise kohta seal, kus neid tegelikult vaja ei ole.

This bill is a clear example of creating

unnecessary juridical norms.

</premise>

/---/

<claim>

Seega oleme seisukohal, et abi vajavaid noori on vaja aidata,

kuid inimene peab ise abi saamiseks pöörduma või keegi hädas

olemisest märku andma.

Therefore, our standpoint is that a young man

who needs an assistance will be assisted but he has to

appeal himself or anybody else has to give a signal

about his difficulties.

</claim>

</argument

The argument in Example 3 weakly attacks the

reason of a supporting argument presented in the

Minister’s report.

Example 4. Argument attacking a reason of the

major claim.

<argument

- - attacking

/---/

Ei ole ka piisavalt argumenteeritud ega suudetud selgitada,

miks on vaja sellisel määral noorte eraellu sekkuda.

It is not enough argued, why it is necessary to

intervene into private life of young people in such a

degree.

</premise>

/---/

<claim>

/---/

Me vaidlustame valitud meetodit ja selle ulatust.

We protest the method and its extent.

</claim>

</argument

The argument in Example 4 similarly to Example

3 weakly attacks the reason of the argument presented

by Minister in his report.

The following Examples (5 and 6) are the

supporting arguments. Both of them are presented as

reactions to previous attacking arguments.

Example 5. A supporting argument.

<argument

- - supporting

<claim>

Sotsiaaldemokraatlik Erakond kindlasti toetab seda eelnõu.

Social democrats definitely support this bill.

</claim>

Selleks, et seda pilootprojekti läbi viia ja et oleks tagatud

andmekaitse, on vaja see seaduses sätestada.

In order to perform this pilot project and to

guarantee the data protection, it is necessary to

legalise it.

</premise>

Me teame ju väga täpselt, et meie riigis on kogu aeg iga

inimene tähtis.

We precisely know that every person in our

country is important every time.

</premise>

Samas me teame, et paljud noored, kes on oma koolitee pooleli

jätnud, kes hooldavad kodus oma vanemaid või kellel on mingi

muu põhjus, on jäänud kõrvale meie ühiskonna rutiinist,

tavapärasest arengust.

We also know that many young people who have

interrupted their education are maintaining their

parents at home or they have another reason to hang

back from our society, our usual development.

</premise>

/---/

</argument

The claim of the argument in Example 5 is

presented before its premises and it coincides with the

major claim. The argument supports the major claim.

Example 6. A supporting argument.

<argument

- - supporting

Selle seadusemuudatuseta võib osa noori jääda aktiivsest

ühiskonnaelust kõrvale, ehkki teenused ja võimalused nende

aitamiseks on olemas.

Without this amendment, some of the young

people will be eliminated from active life although we

have all the means to help them.

ICAART 2020 - 12th International Conference on Agents and Artiﬁcial Intelligence

826

</premise>

/---/

Olgu veel kord üle kinnitatud, et mingiks nuhkimiseks ei lähe

– kõik toimingud saavad olema kooskõlas andmekaitse reeglitega.

It should be stressed once more that no pursuing

will be carried out – all the actions will be done

according to rules of data protection.

</premise>

<claim>

(major claim)

</claim>

</argument

The major claim is not explicitly presented in the

argument. However, it by default coincides with the

major claim. Therefore, the argument supports the

major claim.

4 DISCUSSION

The paper makes the first attempt to annotate and

analyse the arguments in Estonian political discourse.

The empirical material is formed by the verbatim

records of sittings held in the Estonian Parliament. As

a case study, proceeding of the draft act on social care

is considered. Arguments and different relations

between them (attack, support, and rebuttal) are

annotated. In the analysed records, arguments that

support the motion are prevailing over the

counterarguments and the act is approved by the

Parliament.

Every argument consists of two parts – one or

more premises and a claim. In parliamentary

discussions, the presented arguments typically

include more than one premise and they consist of

many sentences. The claims can be explicit or implicit,

by default derived from the premises. In some cases,

MPs in their turns also make proposals/amendments

in addition to rebutting or attacking arguments. The

situation is different as compared with persuasive

essays where premise(s) and a claim are typically

located in the same sentence (Stab, Gurevich 2014).

The inner structure of arguments presented in

parliamentary negotiations needs additional study.

Some tools can be used for (manual) annotation of the

arguments as well as visualising the attack and

support relations between them (e.g. Janier et al.

2014). A challenging further research question is a

comparative study of political argumentation in

Estonian parliament, on one hand, and in other

parliaments, in different political cultures and

different languages, on another hand.

5 CONCLUSIONS

Verbatim records of sittings of the Parliament of

Estonia can be accessed online. For the current study,

an act on social care is chosen as an example. The

draft act passed three readings. The arguments

presented in the process of adopting the act are

annotated. The structure of arguments and the

relations between the arguments are analysed. Some

examples of the arguments are given.

This study is the first step towards automatic

analysis of political arguments in Estonian

parliamentary discussions. The current task is the

development of the annotation scheme and creating a

corpus where arguments are annotated. The automatic

recognition of arguments in Estonian parliamentary

discourse and comparison with other parliaments

remains for the further work.

ACKNOWLEDGEMENTS

This work was supported by institutional research

funding IUT (20-56) of the Estonian Ministry of

Education and Research, by the European Union

through the European Regional Development Fund

(Centre of Excellence in Estonian Studies), and by the

Estonian Research Council grant no. 1226. The

author is also very thankful to anonymous reviewers

for their valuable comments.

REFERENCES

Amgoud, L., Besnard, P., Hunter, A. 2015. Logical

Representation and Analysis for RC-Arguments. In

IEEE 27th International Conference on Tools with

Artificial Intelligence (ICTAI), 104–110.

Atkinson, K., Baroni, P., Giacomin, M., Hunter, A.,

Prakken, H., Reed, C., Simari, G., Thimm, M., Villata,

S. 2017. Towards Artiﬁcial Argumentation. In AI

Magazine, 38(3). https://doi.org/10.1609/aimag.v38i3.-

2704

Bara, J., Weale, A., Bicquelet, A. 2007. Analysing

Parliamentary Debate with Computer Assistance. In

Swiss Political Science Review 13(4): 577–605.

Besnard, P., Hunter, A. 2014. Constructing argument

graphs with deductive arguments: a tutorial. In

Argument & Computation, 5(1), 5-30.

Chesňevar, C., Mcginnis, J., Modgil, S., Rahwan, I., Reed,

C., Simari, G., South, M., Vreeswijk, G., Willmott, S.

2006. Towards an argument interchange format. In The

Knowledge Engineering Review archive, 21(4), 293 –

316.

Arguments in Parliamentary Negotiation: A Study of Verbatim Records

827

Fišer, D., Lenardič, J. 2018. CLARIN resources for

parliamentary discourse research. In ParlaCLARIN

2018 Workshop Proceedings. The Eleventh

International Conference on Language Resources and

Evaluation (LREC 2018), 6 p.

Haddadan, S., Cabrio, E., Serena Villata, S. 2018.

Annotation of Argument Components in Political

Debates Data. In Proceedings of the Workshop on

Annotation in Digital Humanities, 12-16.

Janier, M., Lawrence, J., Reed, C. 2014. OVA+: An

Argument Analysis Interface. In Frontiers in Artificial

Intelligence and Applications, vol, 266, 463-464.

https://doi.org/10.3233/978-1-61499-436-7-463

Koit, M., Õim, H., Roosmaa, T. 2019. How Do the

Members of a Parliament Negotiate? Analysing

Verbatim Records. In Proceedings of the 11th

International Joint Conference on Knowledge

Discovery, Knowledge Engineering and Knowledge

Management (IC3K 2019), 2 (KEOD). Vienna –

Austria: SCITEPRESS – Science and Technology

Publications, Lda, 329−335.

Lippi, M., Torroni, P. 2016a. Argument mining from

speech: detecting claims in political debates. In

Proceedings of the Thirtieth AAAI Conference on

Artiﬁcial Intelligence (AAAI’16), 2979-2985. AAAI

Press.

Lippi, M., Torroni, P. 2016b. MARGOT: A web server for

argumentation mining. In Expert Systems with

Applications, vol. 65, 292-303. https://doi.org/-

10.1016/j.eswa.2016.08.050

Menini, S., Cabrio, E., Tonelli, S., and Villata, S. 2018.

Never retreat, never retract: Argumentation analysis for

political speeches. In The Thirty-Second AAAI

Conference on Artificial Intelligence (AAAI-18), 4889-

4896.

Naderi, N., Hirst, G. 2015. Argumentation mining in

parliamentary discourse. In CMNA 2015, IWEC 2015,

IWEC 2014: Principles and Practice of Multi-Agent

Systems, 16-25.

Petukhova, V., Malchanau, A., Bunt, H. 2015. Modelling

argumentative behaviour in parliamentary debates: data

collection, analysis and test case. In Principles and

Practice of Multi-Agent Systems. CMNA 2015, IWEC

2015, IWEC 2014. Lecture Notes in Computer Science,

vol 9935. Springer, Cham. https://doi.org/10.1007/978-

3-319-46218-9_3

Petukhova, V., Raju, M., Bunt, H. 2017. Multimodal

markers of persuasive speech: designing a Virtual

Debate Coach. In INTERSPEECH, 142-146,

https://doi.org/10.21437/Interspeech.2017-98

Reed, C., Palau, R.M., Rowe, G., Moens, M.-F. Language

Resources for Studying Argument. In Proceedings of

the International Conference on Language Resources

and Evaluation, LREC 2008, 2613- 2618.

Riigikogu (Parliament of Estonia) https://www.riigikogu.-

ee/en/

Stab, C., Gurevych, I. 2014. Annotating Argument

Components and Relations in Persuasive Essays. In

Proceedings of COLING 2014, the 25th International

Conference on Computational Linguistics: Technical

Papers, 1501–1510, Dublin, Ireland.

Venkata, S., Rohit, K., Navjyoti, S. 2018. Analysis of

Speeches in Indian Parliamentary Debates. In CICLING

2018 https://arxiv.org/abs/1808.06834

Vilares, D. 2017. Detecting Perspectives in Political

Debates. In Proceedings of Conference on Empirical

Methods in Natural Language Processing, 1573–1582.

Copenhagen, Denmark.

Walker, M., Tree, J. F., Anand, P., Abbott, R., King, J.

2012) A corpus for research on deliberation and debate.

In Proceedings of the Eighth International Conference

on Language Resources and Evaluation (LREC'12),

812–817.

Working 2017. CLARIN-PLUS Workshop "Working with

Parliamentary Records" https://www.clarin.eu/event/-

2017/clarin-plus-workshop-working-parliamentary-

records

ICAART 2020 - 12th International Conference on Agents and Artiﬁcial Intelligence

828