Latest version: 1.1 [download link]

Previous versions: 1.0

dialign is a software that provides automatic and generic measures of verbal alignment and self-repetitions in dyadic dialogue based on sequential pattern mining at the level of surface of text utterances.

A good place to start can be found in the following paper (more information can be found in the "Citing dialign" section):

• Dubuisson Duplessis, G.; Langlet, C.; Clavel, C.; Landragin, F., Towards alignment strategies in human-agent interactions based on measures of lexical repetitions, Lang Resources & Evaluation, 2021, 36p. [HAL | DOI]

Table of content:

• Framework
• Installation
• Usage
  • Tutorial to use dialign for corpus studies
  • Demonstration of dialign-online for interactive purposes
• Contributors
• Citing dialign
• License

dialign is based on the observation that the behaviours of dialogue participants tend to converge and automatically align at several levels (such as the lexical, syntactic and semantic ones). One consequence of successful alignment at several levels between dialogue participants is a certain repetitiveness in dialogue leading to the development of a lexicon of fixed expressions. As a matter of fact, dialogue participants tend to automatically establish and use fixed expressions that become dialogue routines. More concretely, here follows an excerpt of a dialogue between a human and an agent operated by a Woz where instances of shared lexical patterns are coloured (from the journal article):

dialign provides a framework to quantify the interactive lexical alignment process and the self-repetition behaviour of dialogue participants (DPs) in dyadic textual dialogues. This framework focuses on lexical patterns occurring in dialogue utterances. It distinguishes two main types of such patterns. The first type is shared lexical patterns between DPs, i.e., patterns that are initiated (or primed) by a DP, subsequently adopted by the other DP and possibly reused during the dialogue by any DP. These patterns are directly related to the interactive verbal alignment process, a particular type of on-the-fly linguistic adaptation. They can be seen as shared dialogue routines at the lexical level. They are a way to verbally align and ultimately share a common language to improve understanding, collaboration and social connection to a conversational partner. The second type is lexical self-repetition. Contrary to the previous type which considers patterns that are shared between DPs, self-repetition considers each DP in isolation. Self-repetitions are lexical patterns appearing at least twice in the dialogue utterances of a given DP, independently of the other DP's utterances. Self-repetitions are directly related to the self-consistency of the linguistic production of a given DP.

The main concept behind this model is the automatically built lexicon. For each dialogue transcript, three lexicons are automatically computed:

• the shared expression lexicon: keeps track of shared expressions and valuable features about these expressions (e.g., who first produced this expression, its frequency)
• one self-repetition lexicon per DP: keeps track of self-repetitions and valuable features about these patterns (e.g., its frequency)

Lexicons and the dialogue transcript are leveraged by deriving offline and online measures to quantify aspects of the verbal alignment process and the self-repetition behaviour of DPs. Offline measures are intended to be used for past dialogue interactions (e.g., corpus studies) while online measures are intended for use in a dialogue system.

dialign currently provides out-of-the box offline measures for corpus studies. Online usage in a dialogue system is available as a demonstration.

dialign provides a set of measures to characterise both:

1. the interactive verbal alignment process between dialogue participants, and
2. the self-repetition behaviour of each participant.

These measures allow the characterisation of the nature of these processes by addressing various informative aspects such as their variety, strength, complexity, stability, and orientation. In a nutshell:

• variety: the variety of shared expressions or self-repetitions emerging during a dialogue relative to its length. It is directly related to the number of unique expressions in a lexicon.
• strength: the strength of repetition of the (shared) lexical patterns, i.e., how much the patterns are reused.
• complexity: the complexity indicates the variety of the types of lexical patterns. It is here featured by Shannon entropy measures. High entropy indicates the presence of a wide range of lexical patterns relative to their lengths in number of tokens (e.g., ranging from a single word to a full sentence). On the contrary, low entropy indicates the predominance of one type of lexical pattern.
• extension and stability: The extension and stability of the (shared) lexical patterns are related to the size of the lexical patterns. The extension indicates the size of the lexical patterns. The longer it is, the more extended the lexical pattern is. Extension is directly linked to the stability of the processes since the more extended the patterns are, the more stable the processes are.
• orientation: the orientation of the interactive alignment process, i.e., it indicates either a symmetry (both dialogue participants initiate and reuse the same number of shared lexical patterns), or an asymmetry (a dialogue participant initiates and/or reuses more shared lexical patterns).

(*) All these measures are related to the interactive verbal alignment process

(**) Measures starting with 'S' are related to the self-repetition behaviour, the others are related to the interactivate verbal alignment process

A ready-to-use JAR is available on github. Check the latest release!

You can generate the JAR from SBT.

First, clone the repository. Then, you can compile the code:

$ sbt compile

Eventually, you can produce the JAR as follows (requires sbt-assembly):

$ sbt assembly

The JAR file can be probably found in the directory dialign/target/scala-2.13/.

dialign is designed to be easy to use from the command line interface.

dialign provides out-of-the box offline measures for corpus studies.

A complete walkthrough tutorial is available in the examples/dialign-offline/ directory.

In this tutorial, you will:

• learn how to format your dialogue transcripts in tsv format ;
• learn how to run dialign on a single dialogue transcript (generalisation to a full corpus is straightforward) ; and
• understand the output files of dialign where you can find
  • the measures provided by dialign for the full corpus in tsv format,
  • the shared expression lexicon in tsv format for each dialogue transcript, and
  • the self-repetition lexicons in tsv format for each dialogue transcript.

Let's say that the dialogue files are in the input directory input-directory/ and that output is planned in the directory output-directory/. To run dialign with this configuration, proceed as follows:

java -jar dialign.jar -i input-directory/ -o output-directory/

(here we assume that the dialogue files are encoded in UTF-8, if not it is possible to specify a different encoding by adding -Dfile.encoding=ISO-8859-1 where ISO-8859-1 is the desired encoding)

dialign allows to filter input dialogue files by prefix, suffix and extension. For instance, if the only input dialogue files to consider are files matching the following pattern: dialogue-*-cleaned.dial, it is possible use the following options with dialign:

java -jar dialign.jar -i input-directory/ -o output-directory/ \
	-p "dialogue-" \ # specification of a required filename prefix
	-s "-cleaned" \ # specification of a required filename suffix
	-e "tsv" # specification of the extension (without the '.')

More options are available, see usage note:

java -jar dialign.jar -h

This framework can also be embedded in an interactive system. To demonstrate these capabilities, a complete tutorial is available in the examples/dialign-online/ directory.

In this tutorial, you will:

• learn how to run dialign-online in interactive mode and export the transcript of the created dialogue ; and
• learn how to run dialign-online on a single dialogue transcript in order to directly compute online metrics for each turn.

A screenshot of this demonstration can be found below:

• Guillaume Dubuisson Duplessis (2017, 2020, 2021, 2022)

If you want to refer to the framework or to the dialign software, please cite the following paper:

• Dubuisson Duplessis, G.; Langlet, C.; Clavel, C.; Landragin, F., Towards alignment strategies in human-agent interactions based on measures of lexical repetitions, Lang Resources & Evaluation, 2021, 36p. [HAL | DOI]

If you want to refer to the study strictly limited to verbal alignment on a Human-Agent negotiation task, please cite this paper :

• Dubuisson Duplessis, G.; Clavel, C.; Landragin, F., Automatic Measures to Characterise Verbal Alignment in Human-Agent Interaction, 18th Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL), 2017, pp. 71--81 [See paper | BIB]

The authors of this work would be happy to hear about you if you are using this code! Please, do not hesitate to contact us:

• G. Dubuisson Duplessis <website>
• C. Clavel <website>
• F. Landragin <website>

CECILL-B - see the LICENSE file.