In this paper we summarize our experience and the initial results from implementing and operating IBM Scenario Planning Advisor (SPA), a decision support system that uses lightweight semantic models to assist nance organizations in identifying and managing emerging risk, a category of risk associated with the changes in the global or local economies, politics, technology, society, and others.
SPA is designed to support the business process called \scenario planning"
[
The main functions of SPA are: 1) discovering active risk drivers by aggregating relevant news from the Web and social media, and generating lists of candidate observations corresponding to the detected risk drivers; 2) generating multiple alternative future scenarios highlighting their business implications and leading indicators, based on user-selected observations, and using domain knowledge about driver relations, cascading e ects, and implications.
A key design decision in SPA is that the system does not compute probabilities for the generated scenarios. Instead, we recommend that the domain experts assign probabilities to the nal 3-5 scenarios when necessary. In our experience that approach delivers value to the users much faster and requires signi cantly less work from the experts, compared to creating prediction models for over 200 often non-stationary risk drivers (e.g., oil prices or election results), and their interactions, as would be required to derive scenario probabilities automatically.
We have identi ed the following major challenges in developing SPA: 1) capturing observations from news and social medial; 2) capturing the domain knowledge about the risk drivers quickly and e ciently, while preventing con icts; 3) reasoning with incomplete and biased input to include su ciently complete and minimally biased sets of risks and opportunities in generated scenarios.
In the rest of the paper, we describe the components of the SPA system, explaining their role in addressing the challenges, and highlighting the use of semantic technologies throughout the SPA system. We then present some details of the deployment that indicate our initial success in overcoming the challenges.
Domain Knowledge Workflow
Domain Experts Local Experts
InIHcnrciegraehsaiisnnignflgtartdaiodenbetdleefvieclstdeprCUeucSiradetoinolcnayragaTirnasptedLcoawshCeArodcnomtomseoesiCwdtihecoanrdltkieeffmnyograocinpnedgcoaerpntuivtnairilotanevmsatieEolnactibonlfnveooerasmtCtAbio&cecntmdteoeexsercpolaiantntoeeds
Mind Maps Social Media, RSS/Atom
Customizations AggNreewgsator InHflaigtihoWnEecaokneonminyg
WIKIDATA Observations
AI Planning The above gure shows the interactions between SPA components an d2 its users.
Capturing Observations. The News Aggregator component aggregates
news from RSS and Atom feeds and social media posts, e.g., Twitter, in
multiple languages, by monitoring user-con gured keywords for each candidate
observation, for each country. To further re ne and lter the information, News
Aggregator uses the structured semantic knowledge available in Wikidata.
Country relevance is determined based on the mentions of the local people and
organizations found using Wikidata Query Service [
Capturing Domain Knowledge. The Mind Maps and The
Customizations components store knowledge about risk drivers and business implications
elicited from the domain experts and the local country experts correspondingly.
While the reasoning engine in SPA supports a rich representation of risk drivers
as actions in Planning Domain Description Language (PDDL) [
Reasoning With Incomplete Knowledge. The AI Planning component applies plan recognition and top-k planning techniques to reason with incomplete knowledge and generate scenarios [4{6]. The scenarios are clusters of high-quality plans that include a trajectory of cause-e ect transitions from the Mind Maps, explaining the largest possible subset of observations (rather than achieving a prede ned goal), and such that each plan ends with a business implication. The scenarios, presented as generated text summaries and graphically, are then reviewed and re ned by scenario planning teams.
The deployed system has over 30 active teams of users, 700 scenarios generated, and is processing over 50,000 social media messages per hour. There are 230 risk drivers and business implications and 382 edges between them in the lightweight Mind Map representation of the domain knowledge used for scenario generation. We thank Fang Yuan and Finn McCoole at IBM for providing the domain expertise. We thank Nagui Halim and Edward Shay for their guidance and support. We also thank our LAS collaborators. This material is based upon work supported in whole or in part with funding from the Laboratory for Analytic Sciences (LAS). Any opinions, ndings, conclusions, or recommendations expressed in this material are those of the authors and do not necessarily re ect the views of the LAS and/or any agency or entity of the United States Government.