Pervasive misinformation is a primary social issue in the digital age. A common method for resolving this issue is making corrections to mitigate false beliefs due to misinformation. However, the influence of misinformation is often predominant, thereby resulting in correction having a limited effect on alleviating people's false memory and reasoning. This psychological phenomenon is known as the continued influence effect of misinformation. Rapidly evolving research has accumulated into a sizable literature explaining the psychological processes that cause this effect. This article seeks to clarify the psychological processes for exploring ways to harness the negative impact of misinformation on our minds. Specifically, we review cognitive models and factors related to the continued influence effect, as well as a potential side effect of correction. Moreover, we summarize practical recommendations for interventions based on psychological characteristics. Finally, we discuss future directions in psychology and how emerging interdisciplinary research contributes to controlling the harmful impact of misinformation on our society.

This paper presents a preliminary analysis of an active Bayesian observer that communicates with humans to correct the overconfidence effects due to their confirmatory biases. Two types of confirmatory biases, called Z-and E-types, are introduced and formalized using a model of noisy channels between signals and perceptions. Persons suffering from Z-type confirmatory bias receive a signal and perceive it incorrectly as a different signal with a certain probability, while those suffering from E-type confirmatory bias fail to receive a signal with a certain probability. A dynamic Bayesian network model is developed to analyze the effects of Z-and E-types in a unified way. The analytical model enables us to give a theoretical insight into some basic properties of the active Bayesian observers.

Electronic word-of-mouth (eWOM) is an important information source that influences consumer product evaluations. The author previously developed a computational model, called an inference space model, that predicts potency-magnitude relations between eWOM messages by comparing their numerical values, called inference quantum values. This paper focuses on superlative type messages and formalizes a message differentiation, which classifies them into two kinds: y(*)-open and y(*)-in. The differentiation enables us to investigate the various classes of the messages with respect to their Q-magnitude relations, which are the results of comparing the inference quantum values. As a complement to the previous work, y(*)-in messages are investigated to know how the Q-magnitude relations of y(*)-in messages change depending on message and evaluation parameters. The results show that most of Q-magnitude relations of y(*)-in messages do not depend on the evaluation parameters and suggest that the message differentiation plays a key role to construct accurate prediction strategies based on the inference space model.

Electronic word-of-mouth (eWOM) is an important information source that influence consumer product evaluations. The author previously developed a computational model, called an inference space model, that predicts potency-magnitude relations of eWOM messages involving subjective rank expressions, which refer to the linguistic representations related to the attitude-levels of the benefit of product attributes. This paper focuses on superlative type messages and proposes two effects of eWOM messages on message-receiver's evaluative inferences: a comparison set effect and a difference degree effect. A model and the interpretation of superlative type messages are developed as the basis for formalizing and parameterizing the two effects. This paper considers four groups of message-classes, which are naturally generated from the perspective of the presence or absence of the two effects, to investigate potency-magnitude relations. The investigations include the development of a Q-magnitude Relation Map (Q-Map) that illustrates how the relations change based on the values of two evaluation parameters: evaluation target size and evaluation scale size. The results show that (1) Q-magnitude relations rarely depend on the evaluation parameters and (2) the two effects may be canceled to make the relations indeterminate.

Electronic word-of-mouth (eWOM) is an important information source that influences consumer product evaluations. The author previously developed a computational model, called an inference space model, that predicts potency-magnitude relations of eWOM messages involving subjective rank expressions, which refer to the linguistic representations related to the attitude-levels of the benefits of product attributes. This paper mathematically investigates the potency-magnitude relations of message types differentiating the attitude direction and the strength. The investigations include the developments of a Q-magnitude Relation Map (Q-Map) which illustrates how the relations change based on the values of two parameters: evaluation target size and scale-size balancing indicator. The results show that three scale-classes of bipolar rating scales have a critical role in knowing how the relations change. Based on observations of the Q-Maps, unexplored hypotheses on the potency-magnitude relations are developed with respect to messages involving attitude direction without its strength.

Electronic word-of-mouth (eWOM) is an important information source that influences consumer product evaluations. This paper presents a computational model that predicts the potency-magnitude relations of eWOM messages involving subjective rank expressions, which refer to the linguistic representations related to the attitude-levels of the benefits of the product attributes. The amount of required inference for the message receiver to know the attitude-level through the message is quantified as inference quantum by using inference space, which is characterized by two evaluation parameters: evaluation target size and evaluation scale size. The computational model incorporates the idea of inference quantum into the cognitive hypotheses that were developed to account for the potency differences with reference to the expertise levels -experts or novices -of the message receiver of the products. By applying the computational model to simple eWOM messages, the potency-magnitude relations were observed to depend critically on the values of the message receiver's evaluation parameters. This paper defines three messageclasses, which are also studied in the areas of opinion mining and sentiment analysis, and investigates mathematically how the potency-magnitude relations change based on the values of the evaluation parameters. © 2012 ACADEMY PUBLISHER.

Electronic word-of-mouth (eWOM) is an important information source that influences consumer product evaluations. The author previously developed a computational model, called an inference space model, that predicts the potency-magnitude relations of eWOM messages involving subjective rank expressions, which refer to the linguistic representations related to the attitude-levels of the benefits of product attributes. This paper (1) revises the inference space model so as to evaluate the messages that do not state the degree of the attitude strength but state only the attitude direction, positive or negative, and (2) mathematically investigates the potency-magnitude relations of message types differentiating the attitude direction and the strength. The investigations include the developments of a Q-magnitude Relation Map (Q-Map) which illustrates how the potency-magnitude relations change based on the values of two evaluation parameters: evaluation target size and evaluation scale size. The results are discussed from the viewpoint of eWOM message filtering agents that promote consumer product evaluations. © 2012 IEEE.

Electronic word-of-mouth (eWOM) is an important information source that influences consumer product evaluations. The author previously developed a computational model that predicts the potency-magnitude relations of eWOM messages involving subjective rank expressions, which refer to the linguistic representations related to the attitude-levels of the benefits of product attributes. This paper extends the typology of eWOM messages and investigates mathematically potency-magnitude relations with a focus on intensified comparative expressions, which involve the "degree" of the differences between two or more objects. The investigations include the developments of a Q-magnitude Relation Map (Q-Map) and a Priority Message-Class Map (P-Map), which illustrate how the relations change based on the values of two evaluation parameters: evaluation target size and evaluation scale size. The effects of the messages involving intensified comparative expressions are discussed from the viewpoint of eWOM message filtering agents that promote consumer product evaluations. The investigations were performed first under the premise of a strict assumption called an interval assumption and the results were then verified and discussed under the premise of a conservative assumption called a surjection assumption. © 2012 IEEE.

Electronic word-of-mouth (eWOM) is one important information source that influences consumer product evaluations. This paper presents a computational model that accounts for the potency differences in eWOM messages involving subjective rank expressions, which refer to linguistic representations related to the evaluated levels of the benefits of product attributes. The computational model postulates various types of subjective rank expressions obtained using techniques from opinion mining and sentiment analysis. The computational model incorporates the idea of inference space that contains all possible evaluated levels inferred by the message receiver. The computational model (1) evaluates inference quantum of each message numerically by using the inference space and (2) derives potency-magnitude relations by discerning the expertise level of the message receiver of the products. The unexplored behavioral hypotheses suggested by the computational model are discussed from an analytical viewpoint. © 2011 IEEE.

Electronic word-of-mouth (ewom) is one important information source that influences consumer purchase intentions. Previous works showed that the potency of ewom information, which does not match prior attitudes, tend to be limited. This paper investigates the limitation effects for the ewom information of product attributes from the viewpoint of two layers of consumer prior attitudes: toward products and toward their attributes. The experiment was designed in two parts: (1) subjects evaluated digital cameras using press releases from the makers to form each layer of prior attitudes, and (2) subjects changed their purchase intentions toward the cameras based on the ewom information of the product attributes. A two-way ANOVA with prior attitude toward the products and their attributes as the two factors was performed on the data set from questionnaire surveys of 152 university students. The results supported three hypotheses that reflected the limitation effects of two layers of prior attitudes.

This paper presents an evaluation framework for analytical methods of integrating eWOM Information. This framework involves a communication model that assumes a set of human subjective probabilities called an belief source and includes two translation processes: (1) encoding the belief source into a representation to communicate with a computer; these encoded messages are called e WOM messages, and (2) in the computer, decoding the eWOM messages to estimate the probabilities in the belief source. The efficiency of reducing the difficulty of describing the belief source and the accuracy of reconstructing the belief source are quantitated using this model. The evaluation processes are illustrated with an analytical method of integrating eWOM messages for probabilistic classification problems.

Electronic word-of-mouth (eWOM) is one important information source that influences consumers' product evaluations. This paper presents (1) hypotheses for the potency of eWOM messages with a focus on subjective rank expressions, which refer to linguistic representations related to the level of subjective evaluations and the strength of recommendations, and (2) the results of hypothesis testing on the dataset collected from a questionnaire survey administered to one hundred and fifty two undergraduate students. Two expression types of subjective rank expressions - comparison and degree - were examined. A two-way ANOVA was performed to test the effects of two independent variables "evaluation skill" (SKILL) and "expression type" (TYPE) on the dependent variable "degree of positive change in the evaluations" (POTENCY). The results provide some pieces of evidence in support of the hypotheses. The findings obtained through the research are discussed from a viewpoint of developing accurate methods for the potency prediction of eWOM messages. © 2010 IEEE.

Through an analysis of university students' job-hunting logs, we have found that their introspection via rereading their log sometimes helps them discover themselves. Then we have built a system called PLASIU designed to support job-hunter's creative decision-making based on the observations from their actual job-hunting process. This paper provides an overview of PLASIU and describes the findings from a user study using PLASIU. © Springer-Verlag London Limited 2007.

企業提携における共同経験の役割を分析

This paper describes the impact of WWW on research areas of decision support based on decision support systems which acquire information from the Web. Up-to-date and rich information on the Web will make humans easy to construct decision problems systematically. In order to do so, however, information on the Web must be gathered and be integrated, then be interacted with the users. This paper particularly focuses on the relationship between the Web, data, and users with a viewpoint of human decision processes and decision support systems.

就職活動における気づきの推移を観察し、規則性を抽出

技術移転におけるインターネットコミュニティの役割を論述

The royalty rate for Japanese patents is usually 2%-3% in the computer domain. This rate is based on the royalty rate values described in the "Royalty Rate" (published by Japan Institute of Invention and Innovation) and those used by Japanese national holding patents. This paper demonstrates that this conventional royalty is inadequate for large-scale software application systems and large-scale integrations (LSIs), especially the AP system using business method patents. The reason is that the contribution of the invention is very small for these large-scale systems. Moreover, this paper proposes a new method to calculate the royalty rate for large-scale applications in order to resolve these problems. Experimental case analyses show that the proposed method can derive adequate royalty rates, that pioneer patents have high royalty rates and that bubble patents have low royalty rates.

This paper presents a mathematical model for decision making processes where the knowledge for the decision is constructed automatically from subjective information on the Internet. This mathematical model enables us to know the required degree of accuracy of knowledge acquisition for constructing decision support systems using two technologies: automated knowledge acquisition from information on the Internet and automated reasoning about the acquired knowledge. The model consists of three elements: knowledge source, which is a set of subjective information on the Internet, knowledge acquisition, which acquires knowledge base within a computer from the knowledge source, and decision rule, which chooses a set of alternatives by using the knowledge base. One of the important features of this model is that the model contains not only decision making processes but also knowledge acquisition processes. This feature enables to analyze the decision processes with the sufficiency of knowledge sources and the accuracy of knowledge acquisition methods. Based on the model, decision processes by which the knowledge source and the knowledge base lead to the same choices are given and the required degree of accuracy of knowledge acquisition is quantified as required accuracy value. In order to show the way to utilize the value for designing the decision support systems, the value is calculated by using some examples of knowledge sources and decision rules. This paper also describes the computational complexity of the required accuracy value calculation and shows a computation principle for reducing the complexity to the polynomial order of the size of knowledge sources. Copyright © 2004 JSAI.

Decision Support for Internet Users, called DSIU, is an area of research, we proposed, for decision support systems using two technologies: automated knowledge acquisition from information on the Internet and automated reasoning about the acquired knowledge. This paper describes our research progress and a new research on mathematical models for designing decision support systems in DSIU with respect to realizing the systems and giving contributions to society in the near future. The research progress is described in terms of four automated technologies: (1) judging relevance between words from their meaning, (2) acquiring knowledge from text, (3) handling incomplete and uncertain knowledge, (4) gathering Web-pages using layers of Web-sites, and of two confirmations of the technologies: (5) making check on knowledge in various types of Web-pages and (6) developing prototype systems. The research on mathematical models is described in terms of modeling decision making processeswhere the knowledge for the decision is constructed automatically from information on the Internet. This paper also describes a new challengein terms of building an online shopping mall using some technologies in DSIU. Copyright © 2002 JSAI.

This paper describes availability of personal Web-pages and a prototype development for Decision Support for Internet Users, called DSIU, which is an area of research for decision support by using information on the Internet. The availability of Web-pages concerns usage of formal pages, which are provided by companies and so on, and personal pages, which are provided by private persons. Web-pages are gathered by using an Internet search engine to determine destinations for travel and personal pages are confirmed to provide much subjective information than formal pages. The prototype development concerns a travel recommendation system, which is a kind of decision support systems. The prototype uses subjective and objective information on the Internet to select several destinations for users and to provide explanations the reason why the destinations are recommended. This paper also describes our perspective of DSIU researches. Copyright (c) 2002 JSAI (The Japanese Society for Artificial Intelligence.

仮説推論における仮説の選定に関して新しい考え方を提案

ネット上のテキストから知識獲得する手法を提案

インターネット上の情報の利用について問題点を指摘

For information retrieval (IR) tasks, user models are used to estimate user's true intention and demand. Unfortunately, most user models are constructed in a specialized form that is not applied to other systems or domains. This specialization makes difficulty in sharing user models as common resources for developing information retrieval systems and for researching cognitive characteristics in various users. In order to solve this problem, we need a general user modeling method. In this paper, a user model based on probabilistic framework is proposed. We call this model as generative user model. The generative user model represents user's mental depth by latent (hidden) variables. It also have visible variables that mean word set and qualifier of each word as a subjective probability distribution. The model can handle uncertainty of user's subjectivity by probabilistic framework. Recent statistical studies for such latent models give learning algorithm. Our generative user model can be constructed from dataset taken by information retrieval tasks. As an example, we also introduce two different kinds of information retrieval systems, ART MUSEUM (Multimedia Database with Sense of Color and Construction upon the Matter of ART) and DSIU (Decision Support for Internet Users). The generative user model is applied to these systems. The properties of the model and interactive learning mechanism are shown.

ネット上の情報を使って意思決定支援する枠組みを提案

We are researching decision support systems that acquire knowledge from the Internet. In order to acquire knowledge from text containing daily-used words, it is important to handle a word in terms of its sense. This paper proposes a new semantic word-matching method that manages to acquire knowledge from text containing various daily-used words. We have already constructed a concept base that measures the semantic similarity between daily-used words as a relative meaningful real number between 0 and 1. In this paper, we extend this similarity calculation to deriving an absolute meaningful real number, which enables us to match words according to whether the number is more than or less than a threshold. Using 72 product articles gathered randomly from the Internet, the superiority of the proposed method was confirmed by comparing it with the conventional thesaurus-based matching method.

ネット上のテキストから知識獲得する手法を提案

ネット上のテキストから知識獲得する手法を提案

知識の不確実性を扱う新しい手法を提案

In this paper, we propose an approach to the construction of an intelligent system that handles various domain information provided on the Internet. The intelligent system adopts statistical decision-making as its reasoning framework and automatically constructs probabilistic knowledge, required for its decision-making, from Web-pages. This construction of probabilistic knowledge is carried out using a probability interpretation idea that transforms statements in Web-pages into constraints on the subjective probabilities of a person who describes the statements. In this paper, we particularly focus on describing the basic idea of our approach and on discussing difficulties in our approach, including our perspective.

インターネット上のサービスの技術革新へ向けて、ネットユーザの意思決定支援という視点から、最新の人工知能技術を紹介・解説

本研究では，信頼性の低い情報が多様なデジタル表現で提示されたときに，人がその情報をどのように理解するかを検討し，モデル化と教育への提言を目指す。 21年度は，文献調査の結果の一部について誤情報拡散効果に関するレビュー論文としてまとめた。また，対面で計画していた実験をオンラインで実施し，偽科学的な主張に対してソーシャルネットワークによる他者の意見提示がなされることの影響について検討した。予備的に実施した結果については日本心理学会において学会発表を行った。 予備的実験からは，他者のコメントの価値づけ（Valence:Negative-Positive,メイン情報に賛同するかどうか）の影響がNegativeにおいてのみ見られるということを再現した。ただし，その影響が生じるプロセスについては明確ではなかった。そこで本実験では，Elaboration Likelihood modelを援用し，コメントの価値づけ（Valence:Negative-Positive)と精緻化情報の有無（Elaboration）に注目し，他者のコメントが影響するプロセスをより明確にすることを目的とした。結果からは，主張の内容とは無関係な印象と，内容に関する推論や批判的検討を経て，印象形成と判断，主張への賛同に影響することが示された。先行研究では，主張に対する印象形成や判断に他者の意見が影響するプロセスを「同調」として解釈するものが多いが，本研究では，内容に関する検討のプロセスに提示される意見の質が関連していることを示し，より心理学的に精緻な説明を行うことができた。本実験によって得られた結果については投稿準備中である。 本研究と並行して，異なる提示フォーマットを用いて，他者の意見による批判（Challenge)の焦点が異なることがどのように影響するかを検討する実験を計画し，実験材料の検討・作成を行った。次年度実験を実施し21年度の結果と併せて，デジタル表現における理解・思考により広範に適用できる知見の獲得を目指す。

The inference space model is to quantify the degree of effort required to derive a conclusion from a message as "inference amount" and to explain the attitude change effect among messages from the perspective of the inference amount. The purpose of this research is to experimentally examine the unconfirmed hypotheses derived from the inference space model. Through experiments using fictitious eWOM messages, it was suggested that the messages with a smaller inference amount have a larger attitude change effect.