Chatbot listing, virtual agents, virtual assistants, chat bot directory, conversational agents, virtual human news, chatterbot list
NEWS: Chatbots.org survey on 3000 US and UK consumers shows it is time for chatbot integration in customer service!read more..
Life-Like Characters
Tools, Affective Functions, and Applications
Life-like characters are one of the most exciting technologies for human-computer interface applications today. They convincingly take the roles of virtual presenters, synthetic actors and sales personas, teammates and tutors. A common characteristic underlying their life-likeness or believability as virtual conversational partners are computational models that provide them with affective functions such as synthetic emotions and personalities and implement human interactive behavior. The wide dissemination of life-like characters in multimedia systems, however, will greatly depend on the availability of control languages and tools that facilitate scripting of intelligent conversational behavior.
This book presents the first comprehensive collection of the latest developments in scripting and representation languages for life-like characters, rounded off with an in-depth comparison and synopsis of the major approaches. Introducing toolkits for authoring animated characters further supports practicality and ease of use of this new interface technology.
Life-like characters being a vibrant research area, various applications have been designed and implemented. This book offers coverage of the most successful and promising applications, ranging from product presentation and student training to knowledge integration and interactive gaming. It also discusses the key challenges in the area and provides design guidelines for employing life-like characters.
| Year | 2004 |
| Language | English |
| Publisher | Springer |
| Order: | http://www.amazon.com/Life-Lik... |
| ISBN | 9783540008675 |
| Keywords | life-like characters, animated character |
| This work focuses on the technological developments of life-like characters or embodied agents that live on the screen of computational devices. These embodied agents are being used as virtual tutors, sales persona, guides, advisors, teammates and personal representatives. The editors discuss the issues related to creating successful character-based applications with believable behaviour and characters. The book in general explores various authors discussing the computer programming or scripting languages and tools available for developing these agents, as well as specific prototypes of systems and applications developed. While the scripting languages presented may currently outdated, some of the prototypes and applications developed in this book are still relevant beyond current developments. In the first chapter, the editors define the terms and concepts and provide background on what life character or embodied agents are, how are they used and created as well as why they are being developed. The primary purpose of life-like characters is to improve and make human-computer interaction more meaningful, realistic and natural. Yasuhikoo Kitamura presents an interesting chapter discussing Web information integration using multiple character agents. The author discusses that the Web technology can be very expressive offering the opportunities to use not only text, but images and audio. Kitamura believes that the Web interface can be further enhanced and made more dynamic and interactive through the inclusion of life-like agents. In Kitamura’s chapter, the author proposes a platform for Web information retrieval that uses multiple software agents that collaborate to bring the user information. In Kitamura’s prototype, the use of multiple agent characters refines the ambiguity of a user’s search. This is achieved, by various agents posing questions or making suggestions to users, who can respond to them in dialogue boxes. Apart from helping in refining information sources from the Web, the use of multiple agent characters can also help users to integrate information from various disciplines. Kitamura discusses the prototype called Venus and Mars that consists of three information agents, 1) one that locates recipes from the Web and displays results in the browser, 2) another that locates information from a local knowledge database on cooking ingredients and their relationship to health and 3) one that monitors the input of the user and stores these for future retrieval. These three agents collaborate in order to help a user locate information about food that meet the user’s preferences. Each agent collaborates to locate information for the user through conversation with each other as well as with the user. Kitamura sees such an application as useful for ecommerce and online shopping. Kitamura also sees the application as a useful recommender system in providing information to users who want to locate a restaurant that could meet their specific dietary needs and concerns. Another chapter of relevance is the chapter by Yasuyuki Sumi and Kenji Mase discussing ‘Interface Agents That Facilitate Knowledge Interactions Between Community Members.” In their chapter, the authors document two prototypes, Comic Diary and Agent-Salon. Their prototype of ComicDiary allows people to perform storytelling of their experiences on the web represented in comic style. Agent-Salon on the other hand is said to facilitate face-face knowledge sharing. Both the Agent Salon and Comic Diary represent prototypes that can facilitate exchange of personal expertise among users through visualised information spaces and through conversational stories. Sumi and Mase in their chapter present AgentSalon as a system application of a context-aware mobile assistant that provides personalised tour guidance while facilitating knowledge communications among a community of members with shared interests. In the AgentSalon system, the life-like characters act as media for knowledge interactions between real human beings in a virtual space. AgentSalon is thus created as an environment for conversation among users through the mediation of character agents. The authors before explaining the AgentSalon, differentiate their prototype from previous systems in the literature. They note that other systems in the literature use agents differently mentioning systems that use embodied presentation agents on presentation slides and websites, conversational avatars with gestures in three dimensional spaces and systems that convert texts chats from real online discussion conversations to comics with embodied characters. The authors claim that these existing systems in the literature are based on predefined scripts and do not capture real time conversation. In contrast, the authors’ AgentSalon facilitates encounters and conversations among users, who use the agents to participate in face-to-face conversations. The authors also discuss other systems in the literature where a helper agent provides icebreaking conversational topics for a group of users meeting for the first time. Also mentioned is a topic development agent that is described as providing users with relevant and unexpected topics for discussion by monitoring the conversation and interjecting suggested discussion topics based on retrieval of text from databases. The authors differentiate their system from these agents by arguing that they aim to facilitate new encounters and collaborative knowledge sharing or creation by utilising the personal information of individual users. The authors also review literature on existing knowledge sharing agent systems, mentioning the alter-ego agent that permits asynchronous knowledge sharing, the embodied presentation team of conversational agents, plus a system that generates automatic animation of discussion forums using avatars. The authors again differentiate the AgentSalon system as being different because its knowledge sharing is dynamic and not restricted to previously prepared databases. AgentSalon in contrast by using personal information that is constantly accumulated by the personal agents on mobile hand-held devices (PalmGuides) carried by users, uses real world information to interact in a representational world. AgentSalon is based as a touring guidance system that collects personal information about one’s experience of real world’s tours, as well as interest in particular tours. It detects and represents shared human encounters in the real world and their information interests based on tour records of several users. It then visualizes these different and shared individuals’ viewpoints during online discussion with a semantic map. The authors also discuss and describe the mobile technological device and the portable browser that was used for Agent Salon, which was implemented at a conference. Hence the tour system was based on a conference program of presentations. Users provided feedback about those conference presentations deemed to be of interest to the personal agent via the mobile PalmGuide device, as well as the events that the user actually attended. That information is then represented in AgentSalon. Personal guide agents then migrate from the mobile handheld devices to the AgentSalon platform with users’ personal information and become displayed as animated characters. These agents then share their users’ event attendance records and interests. Agents in the AgentSalon then begin to initiate conversations with the other personal agents, giving users the opportunity to exchange information related to an event. Conversation is then generated (initially by the agents without the users input) based on personal information managed by the agents. Personal information managed by the agents include events visited and the evaluation of those events. Such conversations are generated based on a scripted knowledge base of a reusable set of utterances (called scenes or templates). These scenes are filled in with personal information, such as the user’s name, event title, interest rating or evaluation and visitation time. Agents also discuss shared similarities and differences regarding the experience of an event as well as personal evaluations of an individual’s experience of events. Agents may also recommend events to other personal agents. In the AgentSalon, users can see icons of other users and by clicking on these icons can view more detailed information on events. This feature permits exploration of related information and facilitates deeper discussion by users than that which has been triggered by their personal agents. The authors in the end recommend this application for museums, amusement and theme parks. For Comic Diary, the authors present an interface to allow for the collection and generation of personal stories. In their prototype, a student submitted a comic report on a visit to a museum. This report is displayed in the story form of a comic, structured around the student’s viewpoint. As such the main character in the story represented the person experiencing the museum but with their own identity and personality. In this regard, the main character was the storyteller’s alter ego. In the story, the storyteller is able to exaggerate personal impressions rather than produce actual events. The storytelling permitted the user also to select and highlight impressive events, rather than just recollect all events. In the storytelling performance, the user describes not only the museum’s exhibits but the surroundings and other visitors. The authors conclude that the prototype Comic Diary was an excellent interface for sharing personal impressions and stimulating exchange of individual memories of museum visits, while increasing the motivation to revisit museums. The authors also discuss previous works on artificial intelligent (AI) story generation. They highlight previous works involved developing grammatical schema for generating stories based on analysis of traditional fables, conceptualising problem solving and planning as stories, and developing conversational scripts based on stories. The authors suggest that ComicDiary differs from early AI story generation projects in that they aim to facilitate human conversation or communications through generated stories and represent these stories through comics that will give an overview of personal events. Sumi and Mase also mention research projects that represent chat history as comics or that use captured Web camera images to represent online chat history as a comic. The authors argue that their system translates simple sequence of events into a story and gives users the opportunity to personalise the generated story. |
New Comment
Only registered members are allowed to comment.
or login
Reviews
There is 1 review: