You are hereOverview / Applications
Applications
Technologies based on common sense enable an entirely new breed of applications, ones that are actually "smart" in the sense of understanding the user's situation and goals somewhat like a real person would. Today's machines are sophisticated tools-but tomorrow's will be eager participants in the human world. Imagine...
- a cell phone that, even though you had silenced it in the movie theater, would ring if your mother were to call from the hospital,
- a search engine that, when you entered a gift for my brother, would suggest buying him some new furniture because he had just moved into his first apartment,
- a personal digital assistant that would know to cancel a hiking trip with a friend who had just broken a leg, or
- a camera that, if your sister was running a marathon, would know to take a photo of her crossing the finish line even if you were distracted momentarily.
We have found that common sense can enhance applications in a number of different ways:
- help retrieve related documents and images by expanding queries and annotations using common sense inferencing
-
simplify the interface to expert applications by allowing users to interact with them through common sense rather than specialized language
- provide new dimensions along which to organize and browse media, such as by the emotional dimension
- allow software agents to be more proactive by letting them infer the likely goals of the user and ways to help the user achieve those goals
-
improve context sensing by making use of speech and other sensors to guess the current situation of the user
Projects
To {Do, Go}: ToDoGo uses common sense knowledge and data about local points of interests to associate everyday activities with locations in the Boston area. From this, ToDoGo provides a wide range of applications, including a just-in-time reminder, activity/destination/path recommendations, path optimization, and distributed household management.
SlantExplorer: SlantExplorer is a web-based interface for navigating the conflicting opinions
that underly or are otherwise applicable to a document. It is designed as a
prototypical tool for composing expository documents or for assisting users
trying to summarize long documents for consumption by various groups with
particular interests or backgrounds.
EventMinder: Modern calendar programs do not understand entries, severely limiting their range of assistance. Our prototype calendar application, EventMinder, better understands calendar entries by inferring the components of events and supplying their values.
Storied Navigation:Storied Navigation is a novel approach to constructing a story based on a collection of digital video and audio. Media sequences are tagged with free-text annotations and stored as a collection. The system can then suggest media based on the context of the story.
CommonConsensus: We have developed a fun, self-sustaining web-based game, that both collects and validates common sense knowledge about everyday goals. Goals are a key element of common sense knowledge; in many of our inferface agents, we need to recognize goals from user actions (plan recognition), and generate sequences of actions that implement goals (planning).
Archive
What Am I Gonna Wear?: What Am I Gonna Wear? is a scenario-oriented fashion recommendation system. Electronic commerce on the Web is thriving, but consumers still have trouble finding products to meet their needs and desires. We introduce a novel recommender system technique which works even when users don’t necessarily know exactly what they're looking for.
ARIA: ARIA (Annotation and Retrieval Integration Agent) is a software agent that acts as an assistant to a user writing email or Web pages. As the user types a story, it does continuous retrieval and ranking on a photo database. It can use descriptions in the story text to semi-automatically annotate pictures based on how they are used.
GloBuddy: Globuddy is a common-sense-based application that helps English-speakers to communicate in a foreign language. In addition to translating text, Globuddy provides relevant words and concepts for the traveler's specific situation.
A Common-Sense Approach to Predictive Text Entry: People cannot type as fast as they think, especially when faced with the constraints of mobile devices. The focus of this project is developing an alternative approach to predictive text entry using Open Mind Common Sense.
MakeBelieve:MakeBelieve is a story-generation agent that uses common sense knowledge to interactively compose short fictional texts with a user.
Multilingual Chat for Education: We are developing a new interface for text chat, to assist conversational partners who don't necessarily speak each others' languages perfectly. It will be especially useful to connect students trying to learn foreign languages with native speakers of that language who are trying to learn the first student's language.
Anticipating User Tasks: This project aims to create an agent that, using common sense reasoning, predicts tasks users may be interested in adding to their to-do lists, based on the context of appointments that they are adding to their calendars. For instance, if the user makes the calendar appointment "go fishing," the agent may recommend the task "pack fishing equipment."
EmpathyBuddy: emphasizes collecting common sense facts about the properties and structures of ordinary objects, events, places, goals, and so forth. Some of the activities also collect knowledge in the form of stories and
GloBuddy: emphasizes collecting common sense facts about the properties and structures of ordinary objects, events, places, goals, and so forth. Some of the activities also collect knowledge in the form of stories and
GOOSE: A search engine that uses common sense to reformulate the user's query into a more effective query.
Common Sense in Video Games: When computers automatically generate video game environments, the result is often as simplistic as a randomly generated maze. Using the Open Mind Common Sense knowledge base, we are exploring ways to automatically generate virtual environments that are interesting.