This project aims at developing advanced information access and analysis tools that exploit the strength of collaborative analysis through the interactive support environment where individual users can take advantage not only of the system’s capabilities to rapidly filter and locate relevant information, but also from each other’s actions and insights. The goal is to create an environment where groups of analysts can work together effectively on complex intelligence problems.
For more information, please contact:
Prof. Tomek Strzalkowski, Principal Investigator
University at Albany, SUNY
ILS Room 262B, Social Science Building
Albany, NY 12222.
Email: tomek [at] albany [dot] edu
Phone: 518-442-2608; Fax: 518-442-2606
SUNY Albany is developing COLLANE, a collaborative analytical environment that will allow groups of analysts to work together effectively on complex intelligence problems. Our objective is to deliver significant cognitive augmentation to each analyst by allowing them to share their information and knowledge associated with the case. The system is expected to have measurable operational impact as follows:
- It will allow analysts to work in asynchronous teams on complex tasks against massive data streams including text, internet, and multimedia;
- It will assist analysts in preparing ter quality intelligence less time and with lower cognitive load by combining their findings into a comprehensive product;
- It will help analysts to overcome individual analytical bias by exposing related, contextual and contradictory information, and differences in interpretation;
- It will add cognitive amplification to the process by creating a comprehensive information space reflecting all analysts’ viewpoints and contributions to the task;
- It will facilitate direct and indirect explication and exchange of prior and tacit knowledge and expertise among the analysts.
Our aim in COLLANE is to exploit the strength of collaborative analysis through the interactive support environment where individual analysts can take advantage not only of the system’s capabilities to rapidly filter and locate relevant information, but also from each other’s actions and insights. The goal is to create an environment where the analysts can pursue their individual lines of work while the system makes sure that they are constantly aware of relevant alternative approaches and hypotheses as advanced by other analysts (Figure 1). COLLANE will be designed to allow interoperability between wide varieties of tools that analysts may use.
The technical focus of the COLLANE project is to develop methods and components that will enable collaborative analysis: when they are plugged in, collaborative analysis is enabled. Delivering this vision requires a series of innovations as follows:
- Expand the interactive question answering technology developed for AQUAINT HITIQA so that the system becomes a coordinator and a facilitator in addition to answering direct questions by the analysts. COLLANE will not inhibit the individual lines of inquiry, but it will use its interactive capabilities (both verbal and visual) to communicate similarities and differences among the information requested, collected, retained, and discarded by different analysts.
- Enable COLLANE to automatically compile the Combined Answer Space based on the multiple viewpoints (hypotheses) of an intelligence problem so that the emerging solution can be “rotated” – not unlike a 3-D object – to achieve a full understanding. This will be accomplished by fusing together and semantically correlating the evidence collected by all analysts.
- Investigate a concept of hypothesis footprint, which corresponds to the evidence collected by the analyst along with the internal structure of this evidence (entities, links, events, temporal references, etc.) as well as the steps taken to collect this evidence: questions asked, data viewed, nuggets saved/discarded, etc. In order to construct these, we will exploit and extend Albany’s unique text framing technology to rapidly structure incoming data on the fly. Hypothesis footprints will reveal similarities and mismatches arising between approaches pursued by different analysts.
- Develop a novel multi-channel dialogue management capable of supporting coordinated non-linear interaction between the analysts and the COLLANE information server. COLLANE Dialogue Manager will exploit both verbal (text) and visual communication channels to conduct an efficient dialogue with each user. The system will leverage the multi-threaded dialogue to learn about the case and then turn this knowledge around to amplify each analyst’s cognitive capabilities.
- Prototype a conceptual information visualization for using schematic imagery and detail suppression to convey complex information and dialogue moves such as suggestions, offers, or alerts. The objective is to rapidly communicate the content and complexity of the information accumulated in the case’s Combined Answer Space that reflects the work of multiple analysts. This will provide input to visualization components.
- Provide semantic interoperability with other systems and components so that the information produced by a new component (e.g., geospatial reasoning, image retrieval) can be transparently inserted into the Combined Answer Space where it may be accessed by COLLANE internal modules (credit kim). The objective is to exchange information encapsulated into event frames, which are the elementary building blocks of the Combined Answer Space.
- Handle information in various media and formats, including text, XML, and structured databases. Additional data sources (e.g., imagery, foreign language) can be added with specialized components plugged into COLLANE through semantic interoperability. Semantically interoperable components will produce and accept event frame representations that are handled by COLLANE internal modules.
- View a video demonstration of COLLANE below