Discursis

Discursis

Make sense of big text and visualise conversations automatically.

Discursis is a computer-based tool for analysing human communication.

Communication can be in the form of conversations, web forums, training scenarios, and many more.

Discursis automatically processes transcribed text to show participant behaviours and their interactions around specific topics and over the time-course of the conversation.

Discursis can assist practitioners in understanding the structure, information content, and inter-speaker relationships that are present within input data.

Discursis can:

• track the state of the communication to identify key points and trends and provide early warning of communication issues or changes in situation
• identify whether participants are talking past each other, or engaging successfully
• identify whether participants in the conversation are leaders, promoters, recruiters, innovators, followers, or non-conformists
• identify which concepts are influential, and when.

Discursis presents this data in various visualisations, charts, and reports, including:

• a time series plot of the conversation
• a concept map of the communication content-Discursis uses the Leximancer concept discovery engine to code and map the meaning of the text
• a score card for each conversation participant/channel for behaviour types such as leader, follower, innovator, promoter.

This technology can objectively analyse any textual communication data in real-time, to provide quantitative measures of key behaviours:

• Leadership
• Follower behaviour
• Influence
• Conformity

These measures can be used to:

• Identify whether participants in the conversation are: leaders, promoters, recruiters, innovators, followers, or non-conformists.
• Measure who is being led or influenced by whom, and when, to construct an influence timeline and network.
• Identify which concepts are influential and when.
• Monitor the state of the conversation, to give early warning when the situation is stable, in transition, or chaotic.

Discursis can offer visualisations and reports of all the above features, along with support for the important text data formats:

Data Sources:

• web content – by selected URLs, allowing user configuration of XSL transformation rules for conversation formatting;
• otabular data – dialogue meta-data, such as participant identifier, demographic categories, organisational identifiers, can be supplied by the user as column variables alongside the utterance column;
• files of plain text data – Unicode data; dialogue markup, if present, must follow prescribed format;

Visual Analytic Features:

• Dynamic concept map;
• Real-time state space visualisation, with visual alerting when the conversation enters selected behaviour conditions;
• User configuration of communication channels: these are selected conversation participants, or aggregates of participants, or concept combinations. Each channel is then profiled with a time series chart of its behaviour measures, a summary scorecard, a node on the influence network, and a colour on the recurrence plot;
• Recurrence plot visualisation;
• Dynamic Score Card for each conversation participant/channel for behaviours such as: leader, follower, innovator, promoter;
• Drill-down to concept profiles and supporting text data from all the functions above;
• Automatic generation and visualisation of the Influence Network and the Meme Evolution Tree.

See how Discursis is helping make sense of human communication by reading the latest scientific papers that make use of the technique.

Discursis is available now for use under license.

To get started, choose the relevant license based on whether you are a Student or an Academic.

License fees, terms and conditions do apply, so please read these carefully on the product page.

Payment is via credit card only.

For a whole of faculty/department license, please contact Anne Bannister at UniQuest (a.bannister@uniquest.com.au) for further information.