Intelligent systems with applications, Feb 1, 2023
In this paper we illustrate how novel AI methods can improve the performance of intelligence anal... more In this paper we illustrate how novel AI methods can improve the performance of intelligence analysts. These analysts aim to make sense ofoften conflicting or incompleteinformation, weighing up competing hypotheses which serve to explain an observed situation. Analysts have access to numerous visual analytic tools which support the temporal and/or conceptual structuring of information and collection, and support the evaluation of alternative hypotheses. We believe, however, that there are currently no tools or methods which allow analysts to combine the recording and interpretation of information, and that there is little understanding about how software tools can facilitate the hypothesis formation process. Following the identification of these requirements, we developed the CISpaces (Collaborative Intelligence Spaces) decision support tool in collaboration with professional intelligence analysts. CISpaces combines multiple AI-based methods including argumentation theory, crowdsourced Bayesian analysis, and provenance recording. We show that CISpaces is able to provide support to analysts by facilitating the interpretation of different types of evidence through argumentation-based reasoning, provenance analysis and crowdsourcing. We undertook an experimental analysis with intelligence analysts which highlights three key points. (1) The novel, principled AI methods implemented in CISpaces advance performance in intelligence analysis. (2) While designed as a research prototype, analysts benchmarked it against their existing software tools, and we provide results suggesting intention to adopt CISpaces in analysts' daily activities. (3) Finally, the evaluation highlights some drawbacks in CISpaces. However, these are not due to the technologies underpinning the tool, but rather in its lack of integration with existing organisational standards regarding input and output formats. Our evaluation with intelligence analysts therefore demonstrates the potential impact that an integrated tool building on state-of-the-art AI techniques can have on the process of understanding complex situations, and on how such a tool can help focus human effort on identifying more credible interpretations of evidence.
International Conference on Legal Knowledge and Information Systems, Dec 12, 2018
We demonstrate CISpaces.org, a tool to support situational understanding in intelligence analysis... more We demonstrate CISpaces.org, a tool to support situational understanding in intelligence analysis that complements but not replaces human expertise, for the first time applied to a judicial context. The system combines argumentationbased reasoning and natural language generation to support the creation of analysis and summary reports, and to record the process of forming hypotheses from relationships among information.
This paper briefly describes AIF-EL, an OWL2-EL compliant ontology for the Argument Interchange F... more This paper briefly describes AIF-EL, an OWL2-EL compliant ontology for the Argument Interchange Format.
IEEE Transactions on Visualization and Computer Graphics
Fig. 1. Arigatō prototype and its experiential learning stages for two different topics: Christma... more Fig. 1. Arigatō prototype and its experiential learning stages for two different topics: Christmas and birthday celebration. (a) Reflective observation (RO), (b) Abstract conceptualisation (AC), and (c) Active experimentation (AE), stages for fixed-associations condition on the Christmas celebration topic. (d) Concrete experience (CE) for both topics. (e) Active experimentation (AE), (f) Abstract conceptualisation (AC), and (g) Reflective observation (RO) stages for adaptive-associations condition on a birthday celebration topic. To avoid excessive clutter of the AR scene the black background was selected.
Adaptive Agents and Multi-Agents Systems, May 4, 2015
The aim of intelligence analysis is to make sense of information that is often conflicting or inc... more The aim of intelligence analysis is to make sense of information that is often conflicting or incomplete, and to weigh competing hypotheses that may explain a situation. This imposes a high cognitive load on analysts, and there are few automated tools to aid them in their task. In this paper, we present an agent-based tool to help analysts in acquiring, evaluating and interpreting information in collaboration with others. Agents assist analysts in reasoning with different types of evidence to identify what happened and why, what is credible, and how to obtain further evidence. Argumentation schemes lie at the heart of the tool, and sensemaking agents assist analysts in structuring evidence and identifying plausible hypotheses. A crowdsourcing agent is used to reason about structured information explicitly obtained from groups of contributors, and provenance is used to assess the credibility of hypotheses based on the origins of the supporting information.
We argue that burden of proof (BoP) of the kind present in persuasion does not apply to deliberat... more We argue that burden of proof (BoP) of the kind present in persuasion does not apply to deliberation. We analyze existing computational models showing that in deliberation agents may answer a critique but there is no violation of the protocol if they choose not to. We propose a norm-governed dialogue where BoP in persuasion is modeled as an obligation to respond, and permissions capture the different types of constraint observed in deliberation.
IEEE Transactions on Visualization and Computer Graphics
Fig. 1. VocabulARy prototype. (a) Participant interacting with VocabulARy during the study; (b) V... more Fig. 1. VocabulARy prototype. (a) Participant interacting with VocabulARy during the study; (b) VocabulARy prototype though HoloLens2 in KEYWORD + VISUALISATION instruction mode (the Japanese word "hagaki" sounds as the English phrase "hug a key" (keyword) and visualised with an animated hand grabbing a key (visualisation)); (c) Participant interacting with non-AR version of VocabulARy in KEYWORD instruction mode (Note there is no visualisation of the keyword). AR and non-AR condition were tested with both instruction modes.
Proceedings of the 2019 ACM International Conference on Interactive Experiences for TV and Online Video, 2019
Public, parliamentary and television debates are commonplace in modern democracies. However, deve... more Public, parliamentary and television debates are commonplace in modern democracies. However, developing an understanding and communicating with others is often limited to passive viewing or, at best, textual discussion on social media. To address this, we present the design and implementation of Deb8, a tool that allows collaborative analysis of video-based TV debates. The tool provides a novel UI designed to enable and capture rich synchronous collaborative discussion of videos based on argumentation graphs that link quotes of the video, opinions, questions, and external evidence. Deb8 supports the creation of rich idea structures based on argumentation theory as well as collaborative tagging of the relevance, support and trustworthiness of the different elements. We evaluated the design of the tool in a study of three groups of three people. We present the results of the study and a reflection on the challenges involved.
Restricting the spread of sensitive information is important in domains ranging from commerce to ... more Restricting the spread of sensitive information is important in domains ranging from commerce to military operations. In this position paper, we propose research aimed at exploring techniques for privacy enforcement when humans are the recipient of — possibly obfuscated — information. Such obfuscation could be considered to be a white lie, and we argue that determining what information to share and whether it should be obfuscated must be done through controlled query evaluation, which depends on each agent’s risk/benefit evaluation. We concentrate on machine-human interactions, and note that appropriate specific natural language interfaces need to be developed to handle obfuscation. We propose a solution for creating controlled query evaluation mechanisms based on robust approaches for data representation under uncertainty, viz. SDL-Lite, and propose using ITA Controlled English for natural language generation so as to handle subjectivity. We present the general architecture for our...
In human interactions, trust is regularly updated during a discussion. For example, if someone is... more In human interactions, trust is regularly updated during a discussion. For example, if someone is caught lying, any further utterances they make will be discounted, until trust is regained. This paper seeks to model such behaviour by introducing a dialogue game which operates over several iterations, with trust updates occurring at the end of each iteration. In turn, trust changes are computed based on intuitive properties, captured through three rules. By representing agent knowledge within a preference-based argumentation framework, we demonstrate how trust can change over the course of a dialogue.
Building Information Modelling (BIM) in Design, Construction and Operations, 2015
Building Information Modelling (BIM) in Design, Construction and Operations (ISBN 9781845649142) ... more Building Information Modelling (BIM) in Design, Construction and Operations (ISBN 9781845649142) This version may not include final proof corrections and does not include published layout or pagination.
Proceedings of the 17th International Conference on Distributed Computing and Networking, 2016
Social networks foster the development of social sensing to gather data about situations in the e... more Social networks foster the development of social sensing to gather data about situations in the environment. Making sense of this information is, however, a challenge because the process is not linear and additional sensed information may be needed to better understand a situation. In this paper we explore how two complementary technologies, Moira and CISpaces, operate in unison to support collaboration among human-agent teams to iteratively gather and analyse information to improve situational awareness. The integrated system is developed for supporting intelligence analysis in a coalition environment. Moira is a conversational interface for information gathering, querying and evidence aggregation that supports cooperative data-driven analytics via Controlled Natural Language. CISpaces supports collaborative sensemaking among analysts via argumentation-based evidential reasoning to guide the identification of plausible hypotheses, including reasoning about provenance to explore credibility. In concert, these components enable teams of analysts to collaborate in constructing structured hypotheses with machine-based systems and external collaborators. CCS Concepts •Theory of computation → Automated reasoning; •Human-centered computing → Computer supported cooperative work; Natural language interfaces;
This paper investigates how to restrict the spread of sensitive information. This work is situate... more This paper investigates how to restrict the spread of sensitive information. This work is situated in a military context, and provides a tractable method to decide what semantic information to share, with whom, and how. The latter decision is supported by obfuscation, where a related concept or fact may be shared instead of the original. We consider uncertain information, and utilize Subjective Logic as an underlying formalism to further obfuscate concepts, and reason about obfuscated concepts.
Write Reason is a tool for planning and writing documents. In the left pane is a text editor. In ... more Write Reason is a tool for planning and writing documents. In the left pane is a text editor. In the right pane is a free-form mapping interface. You can link parts of these two panes together, and their contents will stay synchronised.
Models of argument for deliberative Dialogue in Complex Domains: Data Bundle
This data bundle contains all the data that were used in Toniolo's PhD Thesis, with title &qu... more This data bundle contains all the data that were used in Toniolo's PhD Thesis, with title "Models of argument for deliberative Dialogue in Complex Domains". In this research, we defined and empirically evaluated a model of deliberation dialogue based upon argumentation schemes for agents with different objectives to decide what to do in collaboration. The three main folders correspond to the data used for analysis in Chapter 5 (Identifying effective plans through dialogue), Chapter 6a (Strategies driven by plan feasibility) and Chapter 6b (Strategies driven by conflict importance) respectively. Each folder contains files with data resulting from the experiments in csv format, and a file graphAnalysisChX.rtf that contains R scripts for generating the graphs for analysis as presented in the thesis.
Uploads
Papers by alice toniolo