Yasuaki Sakamoto - Academia.edu (original) (raw)
Papers by Yasuaki Sakamoto
Group-level phenomena, such as trends and congestion, are difficult to predict as behaviors at th... more Group-level phenomena, such as trends and congestion, are difficult to predict as behaviors at the group-level typically di- verge from simple aggregates of behaviors at the individual- level. In the present work, we examine the processes by which collective decisions emerge by analyzing how some arti- cles gain vast popularity in a community-based website, Digg. Through statistical analyses and computer
Decision Support Systems, 2014
a b s t r a c t a r t i c l e i n f o Available online xxxx Keywords: Creativity Human based gene... more a b s t r a c t a r t i c l e i n f o Available online xxxx Keywords: Creativity Human based genetic algorithms Advertisement Crowdsourcing Design Evolutionary computing
Computers in Human Behavior, 2014
ABSTRACT Twitter, Facebook, and other social media display the combined opinion of users as colle... more ABSTRACT Twitter, Facebook, and other social media display the combined opinion of users as collective opinion. The purpose of the work reported here was to examine how collective opinion might influence the perceived truthfulness and the sharing likelihood of health-related statements on social media. Experiment 1 revealed that, when evaluating the truthfulness of a statement, participants adopted the collective truthfulness rating associated with the statement. Similarly, Experiment 2 showed that the likelihood that participants would share a statement followed the collective sharing likelihood associated with the statement. These social impacts were extensive, taking place for statements perceived as true, debatable, and false. These results contribute new insights into how people perceive and share information on social media as well as how collective opinion might affect the quality of information on social media.
SSRN Electronic Journal, 2000
ABSTRACT The purpose of the current work is to examine when and how knowing collective opinion in... more ABSTRACT The purpose of the current work is to examine when and how knowing collective opinion influences people’s judgments and decisions in social media environments. In particular, the present work focuses on people’s true-false judgment of statements found on websites and the likelihood of sharing these statements. The results from Experiment 1 revealed that, for false statements, collective opinion had little influence on people’s true-false judgments, but, for true and debatable statements, their judgments were biased toward collective opinion. The results from Experiment 2 indicated that the likelihood of sharing the true, debatable, and false statements followed the collective opinion, and that people were less likely to share false statements than debatable or true ones without collective opinion. These findings extend past work on social influence and advance understanding of how people make judgments and decisions in social media websites.
SSRN Electronic Journal, 2000
2013 46th Hawaii International Conference on System Sciences, 2013
ABSTRACT In this paper, we examined information sharing behavior in social media when one was tak... more ABSTRACT In this paper, we examined information sharing behavior in social media when one was taking the perspective of self versus other. We found that imagining self in a disaster center, Fukushima, Japan, increased the likelihood of sharing crisis information relative to imagining another person, John, in the same place. People's intention to share crisis information by default, without being asked to take any perspective, paralleled the intention to share when taking another person’s perspective. Moreover, when the information was associated with negative feelings, such as worry or fear, it was more likely to be shared; when the information was perceived confusing or uninteresting, it was less likely to be shared.
SSRN Electronic Journal, 2000
ABSTRACT Why do people spread disaster-related news in social media? To address this question, we... more ABSTRACT Why do people spread disaster-related news in social media? To address this question, we analyzed people's tendency to share information discussing the Great East Japan Earthquake and the feelings that they experienced after reading the information in three conditions: when they were asked to think about themselves in a disaster center, when they were asked to think about another person, John, in a disaster center, and when they were not asked to take any perspective. A previous work showed that people who imagined themselves in a disaster center, Fukushima, Japan, were more likely to share related information. We successfully replicated the previous work and extended it by suggesting that feelings could predict the likelihood of information sharing. In this paper, we reported our new findings, proposed a model of information sharing during disaster response, and provided practical implications for advancing the effective use of social media technologies for crises management.
Lecture Notes in Computer Science, 2011
By mapping messages into a large context, we can compute the distances between them, and then cla... more By mapping messages into a large context, we can compute the distances between them, and then classify them. We test this conjecture on Twitter messages: Messages are mapped onto their most similar Wikipedia pages, and the distances between pages are used as a proxy for the distances between messages. This technique yields more accurate classification of a set of Twitter
SSRN Electronic Journal, 2000
2007 IEEE Intelligence and Security Informatics, 2007
Probabilities of physical attack are often determined by various environmental factors. As the en... more Probabilities of physical attack are often determined by various environmental factors. As the environment changes, the probability of attack associated with an area changes. In such dynamic environments, autonomous sensors are potentially useful to optimally cover regions that have high probabilities of attack. We present results from agent-based simulations, in which autonomous sensors "forage" a space to find areas with
SSRN Electronic Journal, 2000
Crowds can be used to generate and evaluate design solutions. To increase a crowdsourcing system'... more Crowds can be used to generate and evaluate design solutions. To increase a crowdsourcing system's effectiveness, we propose and compare two evaluation methods, one using five-point Likert scale rating and the other prediction voting. Our results indicate that although the two evaluation methods correlate, they have different goals: whereas prediction voting focuses evaluators on identifying the very best solutions, the rating focuses evaluators on the entire range of solutions. Thus, prediction voting is appropriate when there are many poor quality solutions that need to be filtered out, and rating is suited when all ideas are reasonable and distinctions need to be made across all solutions. The crowd prefers participating in prediction voting. The results have pragmatic implications, suggesting that evaluation methods should be assigned in relation to the distribution of quality present at each stage of crowdsourcing.
SSRN Electronic Journal, 2000
ABSTRACT During a disaster, information spreads through social media. Although information that c... more ABSTRACT During a disaster, information spreads through social media. Although information that can save lives exists in social media, it can be difficult to find because it is buried under a sea of unverified information. One challenge in improving the use of social media for disaster management is to facilitate the spread of actionable information that improves people’s well-being and at the same time reduce the spread of misinformation that confuses people and interferes with the discovery of useful information. To this end, we study the psychology of message forwarding in social media. Based on the understanding of the psychology, we develop recommendations for designing tools that augment the forwarding decisions of individuals such that useful information spreads in social media. Here, we present the results from analyzing the forwarding of tweets related to the disasters caused by the Great East Japan Earthquake in 2011. The results from questionnaires suggest that people will be more likely to share a disaster-related message in a social media environment when they perceive the message as more important, accurate, anxiety provoking, familiar, informative, and fluent. The implications of the results on improving the quality of information in social media during disaster response are discussed.
SSRN Electronic Journal, 2000
ABSTRACT During and after a disaster, victims and others often take to social media sites to shar... more ABSTRACT During and after a disaster, victims and others often take to social media sites to share information about conditions, aid, resources and the like. But well-intentioned users can spread rumors that are later found to be false, as they did following the 2011 Great East Japan earthquake, which hampered rescue operations and confused people. To improve the quality of information on social media, we study methods for integrating information provided by crowds in social media environments. In this paper, we review some results from our research showing that crowdsourced critical-thinking and veracity evaluation can be effective in curbing the spread of false information on social media. These findings suggest that crowds can help triage information in order to support the discovery of relevant information on social media during and after emergencies.
SSRN Electronic Journal, 2000
2013 International Conference on Social Computing, 2013
ABSTRACT This paper proposes a hybrid human machine system based on an expert weighting algorithm... more ABSTRACT This paper proposes a hybrid human machine system based on an expert weighting algorithm that combines the responses of both humans and machine learning algorithms. The general topic of the paper is the use of the crowd to interpret text, and the power of that interpretation to predict future events. This topic is addressed through an experiment, in which news sentiment is evaluated by crowds and experts in different configurations. Their classifications are used as training sets for machine learning algorithms, including one that weights both machine and human predictions. The testing is done based on Thomson Reuters news stories and the returns of the stocks mentioned right after the stories appear. The hybrid expert weighting algorithm forecasts asset returns similar to the different versions of the trained and crowd groups because it combines the best results of the machine learning algorithms with human answers. The forecast of the expert weighting algorithm does not always show the best performance in comparison with the other learning algorithms, however its performance is very similar to the best algorithm in most cases. From a cognitive perspective, the capacity of the expert weighting algorithm to select dynamically the best expert according to its previous performance is consistent with an evolving collective intelligence: the final decision is a combination of the best individual answers - some of these come from machines, and some from humans.
Journal of experimental psychology. General, 2004
The results from 3 category learning experiments suggest that items are better remembered when th... more The results from 3 category learning experiments suggest that items are better remembered when they violate a salient knowledge structure such as a rule. The more salient the knowledge structure, the stronger the memory for deviant items. The effect of learning errors on subsequent recognition appears to be mediated through the imposed knowledge structure. The recognition advantage for deviant items extends to unsupervised learning situations. Exemplar-based and hypothesis-testing models cannot account for these results. The authors propose a clustering account in which deviant items are better remembered because they are differentiated from clusters that capture regularities. The function of clusters is akin to that of schemas. Their results and analyses expose connections among research in category learning, schemas, stereotypes, and analogy.
SSRN Electronic Journal, 2000
ABSTRACT We examined the benefits of different search strategies by testing four computational mo... more ABSTRACT We examined the benefits of different search strategies by testing four computational models. In one model, agents in a group always innovated. The other three models incorporated some mechanisms of imitation. In the second model, each agent imitated the best solution of a random other. In the third model, each agent followed preferential attachment and imitated the best solution of the agent that was asked by many agents. In the fourth model, each agent developed a familiarity with an agent based on how often it asked a certain agent, and imitated this agent. In two simulation studies, following the most popular or the most familiar agent resulted in a good compromise between efficiency and diversity in finding good solutions. People’s desire to follow particular individuals may be a key to their adaptive behavior, allowing them to disseminate ideas efficiently while encouraging the exploration of new ideas.
Group-level phenomena, such as trends and congestion, are difficult to predict as behaviors at th... more Group-level phenomena, such as trends and congestion, are difficult to predict as behaviors at the group-level typically di- verge from simple aggregates of behaviors at the individual- level. In the present work, we examine the processes by which collective decisions emerge by analyzing how some arti- cles gain vast popularity in a community-based website, Digg. Through statistical analyses and computer
Decision Support Systems, 2014
a b s t r a c t a r t i c l e i n f o Available online xxxx Keywords: Creativity Human based gene... more a b s t r a c t a r t i c l e i n f o Available online xxxx Keywords: Creativity Human based genetic algorithms Advertisement Crowdsourcing Design Evolutionary computing
Computers in Human Behavior, 2014
ABSTRACT Twitter, Facebook, and other social media display the combined opinion of users as colle... more ABSTRACT Twitter, Facebook, and other social media display the combined opinion of users as collective opinion. The purpose of the work reported here was to examine how collective opinion might influence the perceived truthfulness and the sharing likelihood of health-related statements on social media. Experiment 1 revealed that, when evaluating the truthfulness of a statement, participants adopted the collective truthfulness rating associated with the statement. Similarly, Experiment 2 showed that the likelihood that participants would share a statement followed the collective sharing likelihood associated with the statement. These social impacts were extensive, taking place for statements perceived as true, debatable, and false. These results contribute new insights into how people perceive and share information on social media as well as how collective opinion might affect the quality of information on social media.
SSRN Electronic Journal, 2000
ABSTRACT The purpose of the current work is to examine when and how knowing collective opinion in... more ABSTRACT The purpose of the current work is to examine when and how knowing collective opinion influences people’s judgments and decisions in social media environments. In particular, the present work focuses on people’s true-false judgment of statements found on websites and the likelihood of sharing these statements. The results from Experiment 1 revealed that, for false statements, collective opinion had little influence on people’s true-false judgments, but, for true and debatable statements, their judgments were biased toward collective opinion. The results from Experiment 2 indicated that the likelihood of sharing the true, debatable, and false statements followed the collective opinion, and that people were less likely to share false statements than debatable or true ones without collective opinion. These findings extend past work on social influence and advance understanding of how people make judgments and decisions in social media websites.
SSRN Electronic Journal, 2000
2013 46th Hawaii International Conference on System Sciences, 2013
ABSTRACT In this paper, we examined information sharing behavior in social media when one was tak... more ABSTRACT In this paper, we examined information sharing behavior in social media when one was taking the perspective of self versus other. We found that imagining self in a disaster center, Fukushima, Japan, increased the likelihood of sharing crisis information relative to imagining another person, John, in the same place. People's intention to share crisis information by default, without being asked to take any perspective, paralleled the intention to share when taking another person’s perspective. Moreover, when the information was associated with negative feelings, such as worry or fear, it was more likely to be shared; when the information was perceived confusing or uninteresting, it was less likely to be shared.
SSRN Electronic Journal, 2000
ABSTRACT Why do people spread disaster-related news in social media? To address this question, we... more ABSTRACT Why do people spread disaster-related news in social media? To address this question, we analyzed people's tendency to share information discussing the Great East Japan Earthquake and the feelings that they experienced after reading the information in three conditions: when they were asked to think about themselves in a disaster center, when they were asked to think about another person, John, in a disaster center, and when they were not asked to take any perspective. A previous work showed that people who imagined themselves in a disaster center, Fukushima, Japan, were more likely to share related information. We successfully replicated the previous work and extended it by suggesting that feelings could predict the likelihood of information sharing. In this paper, we reported our new findings, proposed a model of information sharing during disaster response, and provided practical implications for advancing the effective use of social media technologies for crises management.
Lecture Notes in Computer Science, 2011
By mapping messages into a large context, we can compute the distances between them, and then cla... more By mapping messages into a large context, we can compute the distances between them, and then classify them. We test this conjecture on Twitter messages: Messages are mapped onto their most similar Wikipedia pages, and the distances between pages are used as a proxy for the distances between messages. This technique yields more accurate classification of a set of Twitter
SSRN Electronic Journal, 2000
2007 IEEE Intelligence and Security Informatics, 2007
Probabilities of physical attack are often determined by various environmental factors. As the en... more Probabilities of physical attack are often determined by various environmental factors. As the environment changes, the probability of attack associated with an area changes. In such dynamic environments, autonomous sensors are potentially useful to optimally cover regions that have high probabilities of attack. We present results from agent-based simulations, in which autonomous sensors "forage" a space to find areas with
SSRN Electronic Journal, 2000
Crowds can be used to generate and evaluate design solutions. To increase a crowdsourcing system'... more Crowds can be used to generate and evaluate design solutions. To increase a crowdsourcing system's effectiveness, we propose and compare two evaluation methods, one using five-point Likert scale rating and the other prediction voting. Our results indicate that although the two evaluation methods correlate, they have different goals: whereas prediction voting focuses evaluators on identifying the very best solutions, the rating focuses evaluators on the entire range of solutions. Thus, prediction voting is appropriate when there are many poor quality solutions that need to be filtered out, and rating is suited when all ideas are reasonable and distinctions need to be made across all solutions. The crowd prefers participating in prediction voting. The results have pragmatic implications, suggesting that evaluation methods should be assigned in relation to the distribution of quality present at each stage of crowdsourcing.
SSRN Electronic Journal, 2000
ABSTRACT During a disaster, information spreads through social media. Although information that c... more ABSTRACT During a disaster, information spreads through social media. Although information that can save lives exists in social media, it can be difficult to find because it is buried under a sea of unverified information. One challenge in improving the use of social media for disaster management is to facilitate the spread of actionable information that improves people’s well-being and at the same time reduce the spread of misinformation that confuses people and interferes with the discovery of useful information. To this end, we study the psychology of message forwarding in social media. Based on the understanding of the psychology, we develop recommendations for designing tools that augment the forwarding decisions of individuals such that useful information spreads in social media. Here, we present the results from analyzing the forwarding of tweets related to the disasters caused by the Great East Japan Earthquake in 2011. The results from questionnaires suggest that people will be more likely to share a disaster-related message in a social media environment when they perceive the message as more important, accurate, anxiety provoking, familiar, informative, and fluent. The implications of the results on improving the quality of information in social media during disaster response are discussed.
SSRN Electronic Journal, 2000
ABSTRACT During and after a disaster, victims and others often take to social media sites to shar... more ABSTRACT During and after a disaster, victims and others often take to social media sites to share information about conditions, aid, resources and the like. But well-intentioned users can spread rumors that are later found to be false, as they did following the 2011 Great East Japan earthquake, which hampered rescue operations and confused people. To improve the quality of information on social media, we study methods for integrating information provided by crowds in social media environments. In this paper, we review some results from our research showing that crowdsourced critical-thinking and veracity evaluation can be effective in curbing the spread of false information on social media. These findings suggest that crowds can help triage information in order to support the discovery of relevant information on social media during and after emergencies.
SSRN Electronic Journal, 2000
2013 International Conference on Social Computing, 2013
ABSTRACT This paper proposes a hybrid human machine system based on an expert weighting algorithm... more ABSTRACT This paper proposes a hybrid human machine system based on an expert weighting algorithm that combines the responses of both humans and machine learning algorithms. The general topic of the paper is the use of the crowd to interpret text, and the power of that interpretation to predict future events. This topic is addressed through an experiment, in which news sentiment is evaluated by crowds and experts in different configurations. Their classifications are used as training sets for machine learning algorithms, including one that weights both machine and human predictions. The testing is done based on Thomson Reuters news stories and the returns of the stocks mentioned right after the stories appear. The hybrid expert weighting algorithm forecasts asset returns similar to the different versions of the trained and crowd groups because it combines the best results of the machine learning algorithms with human answers. The forecast of the expert weighting algorithm does not always show the best performance in comparison with the other learning algorithms, however its performance is very similar to the best algorithm in most cases. From a cognitive perspective, the capacity of the expert weighting algorithm to select dynamically the best expert according to its previous performance is consistent with an evolving collective intelligence: the final decision is a combination of the best individual answers - some of these come from machines, and some from humans.
Journal of experimental psychology. General, 2004
The results from 3 category learning experiments suggest that items are better remembered when th... more The results from 3 category learning experiments suggest that items are better remembered when they violate a salient knowledge structure such as a rule. The more salient the knowledge structure, the stronger the memory for deviant items. The effect of learning errors on subsequent recognition appears to be mediated through the imposed knowledge structure. The recognition advantage for deviant items extends to unsupervised learning situations. Exemplar-based and hypothesis-testing models cannot account for these results. The authors propose a clustering account in which deviant items are better remembered because they are differentiated from clusters that capture regularities. The function of clusters is akin to that of schemas. Their results and analyses expose connections among research in category learning, schemas, stereotypes, and analogy.
SSRN Electronic Journal, 2000
ABSTRACT We examined the benefits of different search strategies by testing four computational mo... more ABSTRACT We examined the benefits of different search strategies by testing four computational models. In one model, agents in a group always innovated. The other three models incorporated some mechanisms of imitation. In the second model, each agent imitated the best solution of a random other. In the third model, each agent followed preferential attachment and imitated the best solution of the agent that was asked by many agents. In the fourth model, each agent developed a familiarity with an agent based on how often it asked a certain agent, and imitated this agent. In two simulation studies, following the most popular or the most familiar agent resulted in a good compromise between efficiency and diversity in finding good solutions. People’s desire to follow particular individuals may be a key to their adaptive behavior, allowing them to disseminate ideas efficiently while encouraging the exploration of new ideas.