Tomonari Kamba

Unintrusive customization techniques for Web advertising

Most online advertisement systems in place today use the concept of consumer targeting: each user is identified and, according to his or her system setup, browsing habits and available off-line information, categorized in order to customize the advertisements for highest user responsiveness. This constant monitoring of a users online habits, together with the trend to centralize this data and link it with other databases, continuously nurtures fears about the growing lack of privacy in a networked society. In this paper, we propose a novel technique of adapting online advertisement to a users short term interests in a non-intrusive way. As a proof-of-concept we implemented a dynamic advertisement selection system able to deliver customized advertisements to users of an online search service or Web directory. No user-specific data elements are collected or stored at any time. Initial experiments indicate that the system is able to improve the average click-through rate substantially compared to random selection methods.

ANATAGONOMY: a personalized newspaper on the World Wide Web

Abstract This paper describes a personalized newspaper on the World Wide Web (WWW), called ANATAGONOMY. The main feature of this system is that the newspaper is personalized without asking the users to specify their preferences explicitly. The system monitors user operations on the articles and reflects them in the user profiles. Differently from conventional newspapers on the WWW, our system sends an interaction agent implemented as a Java applet to the client side, and the agent monitors the user operations and creates each users newspaper pages automatically. The server side manages user profiles and anticipates how interesting an article would be for each user. The interaction agent on the client side manages all the user interactions, including the automatic layout of pages. Our system has page multiple layout algorithms and the user can switch from one view to another anytime, according to the preference or machine environment. On one of the views, the user can even see all the articles sequentially without performing any operations. We evaluated a scheme in which the user scores each article explicitly, and a scheme in which all the personalization is done automatically. The results show that automatic personalization works well when some parameters are set properly.

Learning personal preferences on online newspaper articles from user behaviors

Abstract This paper discusses methods by which user preferences for WWW-based newspaper articles can be learned from user behaviors. Two modes of inference were compared in an experiment: one using explicit feedback and the other using implicit feedback. In the explicit feedback mode, the users score all articles according to their relevance. In the implicit feedback mode, the user reads articles by performing scrolling and enlarging operations, and the system infers from the operations how much the user was interested in each article. Our newspaper on the WWW, called ANATAGONOMY, has a learning engine and a scoring engine on the server. The system users read daily news articles by using a WWW browser in which there is an interaction agent that monitors the user behaviors. The learning engine on the server infers user preferences from the interaction agent, and the scoring engine scores new articles and creates personalized newspaper pages based on the extracted user profiles. In an experiment, the system was able to personalize the newspaper to some extent when using only implicit feedback when some parameters were properly set, but the personalization was not as precise as it was when explicit feedback was used. By mixing explicit feedback with implicit feedback, the system could personalize newspapers quickly and precisely without requiring too much effort on the part of the users. User preferences can also be used to construct information retrieval agents or even to create cyberspace communities of the users that have similar interests. We think that the proposed technique for learning user preferences greatly enhances the value of the WWW.

Effective personalization of push-type systems—visualizing information freshness

Abstract This paper discusses an effective personalization method, especially on push-type systems. Many conventional personalization systems rely strictly on personal interests during information presentation, but the “freshness” of information is often as important as the relation to personal preferences. For example, a user who accesses a WWW newspaper several times a day, expects to see fresh articles displayed in prominent positions rather than “hidden” among articles that may be more relevant but that have already been read. This paper therefore presents a novel personalization method incorporating “information freshness” and that is extremely useful for the ever-growing number of push-type systems. Information freshness is indicated by using a perspective representation which shows virtual depth on the screen: fresh articles seem “closer” to the user, while old articles seem farther away. This representation allows us to simultaneously display both the personal relevance and the freshness of the information. We have successfully implemented two applications using this technique: a personalized newspaper service and an easy-to-use scrapbook for Web pages.

Personalized advertisement-duration control for streaming delivery

This paper describes the development of a streaming advertisement delivery system that controls the insertion of streaming advertisements into streaming content.Conventional personalization techniques lack a time-control function for advertisement insertion, so the advertisement exposure for each user access can become excessive, much to the annoyance of viewers. This could devalue streaming content by making it less attractive.In our technique, advertisement insertion control is based on the history of each viewer. This personalization method makes it possible to maintain a balanced ratio of the advertisement length to the content length. As a result, our technique should encourage the growth of Internet streaming services and enable more effective and less intrusive advertising.

A domain cluster interface for WWW search

Because of the recent explosive increase in the number of WWW documents, directory services are becoming indispensable. In the keyword search function of most directory services, search results are displayed as a URL list ordered by importance as calculated by the system, but the order sometimes does not have any meaning to the user since the calculation algorithm is a black box. In addition, it is difficult to find useful documents from a long list. To solve this problem, the authors have developed a new WWW search system that clusters the documents in the search results by organization name, derived from its URL domain name. The system displays the clusters in a hierarchical tree view form.

A shopping assistant agent for Web-shops

One of the most important problems of online shops on the World Wide Web is that a site visitor finds it difficult to choose an item to buy unless he or she has sufficient knowledge of that item. In this paper, the authors describe an intelligent software agent that works as a shopping assistant. The agent has knowledge about the merchandise and helps visitors to the shop choose items. The interaction with this agent has three features: (1) two interaction channels (selection and conversation); (2) flexible topic change (the user can trigger a new conversation flow even in the middle of a conversation); and (3) personalized interaction (the interaction is personalized according to user behavior). This agent has been applied to our Web site to help potential buyers of built-to-order (BTO) PCs. Implementation details and user behavior on the above site are described.

Presentation of personalized information using anthropomorphous agents

This paper proposes a method to present personalized information effectively using multiple anthropomorphous agents that know the users preferences. Conventionally, techniques such as filtering and sorting are used to show the information customized for each user, but it is difficult to naturally reflect human multi-dimensional preferences in such a presentation format. In the proposed method, each agent has a specific viewpoint and interactively points at the contents that the user will be interested in. This technique has been applied to an Internet-based information service for registered PC users.