This paper introduces a method to transform technical product descriptions into user-friendly experiential descriptions, while also highlighting relevant parts of the original description. Product descriptions often are hard to understand without prior knowledge. For example, a beginner with a camera cannot understand technical descriptions like “ISO sensitivity 51,200”. Our method translated this description to more relatable phrases such as “captures clear faces even at night.” Our method adopts a generative language model to enable such experiential description generation and evidence estimation. Our method first trains a model with pairs of product descriptions and reviews. The trained model generates many candidate experiential descriptions when given product descriptions. After training, our method uses an ablation-based approach to estimate the evident description of the generated candidates. It checks for the frequency of words in the generated narrative when a portion of the description is removed. For example, terms like “night” or “clear” became less prevalent in reviews when “ISO sensitivity” was removed from the input description. Subject experiments with the actual review dataset verified our method's effectiveness in generating accurate narratives highlighting product features.

Web advertising services have exhibited consistent growth over the years. However, the conventional methods of web advertising recommendations, relying on keyword matching with search queries and browsing histories, encounter challenges when it comes to effectively targeting users with hidden or latent interests. In contrast, the use of mobile device location data in advertising recommendations often centers around physical store proximity. To address these limitations, our research aims to enhance web advertising recommendations by analyzing latent user interests through real-world behavioral data. This study specifically investigates the influence of area size on user behavioral analysis and its subsequent impact on the accuracy of predicting visit probabilities. We achieve this by extracting the user’s activity range from user behavior (movement) log data and geotagged tweets. Subsequently, we tally the places visited by the user, considering spot attributes, and convert this data into feature vectors. Utilizing these feature vectors in conjunction with various classification methods, we build learning models. In this paper, we present and evaluate these learning models employing different area sizes, verifying their accuracy in predicting user visits to specific stores.

In recent years, the number of people running to improve their health and physical fitness has been increasing. However, it is not easy to stay motivated and keep running as regular activity. Therefore, we believe that it is very significant to develop a running support system. In a previous study, a running assistance system that enables users to run with a virtual runner based on their own running records in an acoustic augmented reality space was constructed. In order to make the running experience more realistic and keep users motivated it is important to develop a system that allows users to compete with virtual runners who can serve as good rivals. In this study we examined how multiple virtual runners competing against the user affect the user's motivation.

In recent years, the number of people who run for the purpose of improving their health and physical fitness has been increasing, but it is not easy to continue running. Therefore, we believe that it is very important to develop a running support system. In our previous study, we developed a running support system and an application that enables users to run with a virtual runner created by using their past running data in an acoustic augmented reality space. However, this system has the problem that it can only race against the past running records of oneself or one's acquaintance, and can only race against a single running record. Therefore, we considered it necessary to develop a system that allows users to race against an unspecified number of users and that allows users to arbitrarily select the distance and number of times they wish to race. In this paper, we propose an online marathon system that enables large-scale online races and examine the effect of the number of competitors on user motivation.

Web advertising services for the Internet are growing rapidly. However, it is difficult to effectively recommend Web advertisements to latent buyers using the current mainstream Web ad recommendation method based on keyword matching with online search and browser history. This is because it is an explicit analysis using keywords that the user is already interested in. On the other hand, most of the advertisement recommendation methods using location information of mobile terminals are based on the distance to the actual store. Therefore, this research proposes and verifies a method to analyse latent user interest based on analysis of real-space user behavior considering point-of-interest (POI) attributes. We apply the method to Web advertisement recommendation on mobile terminals, and assume that the user’s interest targets are stores that exist in real space. Specifically, we use geotagged tweet data to extract the users’ movement activity range within a certain time period with reference to the store locations. The movement activity range serves as basis for POI attribute analysis using OpenStreetMap (OSM) data. Finally, the characteristics of the real-space movement activity history are extracted and used together with the POI attributes to train a XGBoost model. In previous research, we accumulated data, generated training models with various settings, and verified their prediction accuracy of users visiting a target stores. In this paper, we focus on verifying the effectiveness of the proposed advertising recommendation method by conducting a questionnaire survey of actual users.

We propose a method for learning entity orders, for example, safety, popularity, and livability orders of countries. We train linear functions by using samples of ordered entities as training data, and attributes of entities as features. An example of such functions is f(Entity) $$= +0.5$$ (Police budget) $$-0.8$$ (Crime rate), for ordering countries in terms of safety. As the size of training data is typically small in this task, we propose a machine learning method referred to as context-guided learning (CGL) to overcome the over-fitting problem. Exploiting a large amount of contexts regarding relations between the labeling criteria (e.g. safety) and attributes, CGL guides learning in the correct direction by estimating a roughly appropriate weight for each attribute by the contexts. This idea was implemented by a regularization approach similar to support vector machines. Experiments were conducted with 158 kinds of orders in three datasets. The experimental results showed high effectiveness of the contextual guidance over existing ranking methods.

Background Hepatic artery thrombosis can lead to graft loss associated with severe hepatic infarction or bile duct ischemia. When anatomical hepatic artery reconstruction is impossible in liver transplantation or hepato-pancreatic biliary surgery, portal vein arterialization (PVA) is proposed as a salvage technique. Herein, we report our experience with a case that showed favorable clinical outcomes after partial PVA during living-donor liver transplantation (LDLT) because of difficulties in arterial reconstruction. Case presentation A 62-year-old woman with non-B, non-C liver cirrhosis complicated with hepatocellular carcinoma was being prepared for LDLT using an extended left lobe graft. The graft presented with two arteries (left hepatic artery, 2 mm; middle hepatic artery, 2 mm). The first anastomosis was performed using the recipient hepatic artery stumps, but no flow was detected on Doppler control because of thrombus formation. The next attempt was executed using the middle colic artery with a radial artery jump graft and the right gastroepiploic artery, but it led to the same result. Thus, the graft oxygen support by the standard arterial procurement was abandoned, and a shunt was created between the ileocecal artery and the vein to obtain PVA. Arteriography of the superior mesenteric artery showed that the shunt was relatively patent, and the portal vein was apparent. No biliary complication or liver abscess occurred postoperatively, and the patient presented with good liver function and no complications related to portal vein hypertension, nor liver fibrosis 18 months after the LDLT. Conclusion Partial PVA with a shunt created between the ileocecal artery and the vein is useful when arterial reconstruction is difficult during LDLT for preventing graft loss caused by severe hepatic infarction or bile duct ischemia.

We address the problem of searching for microblogs referring to events, which are difficult to find because microblogs may refer to events without using event’s contents and a searcher may not use suitable queries for a search engine. We therefore propose a dynamic search process based on MDP that takes query strategies optimized for the current search state. As key components of the dynamic search process, we propose an RNN-based model for predicting long-term returns of a search process, and a DNN-based model that tries to match between the representations of microblogs and those of events for identifying relevant microblogs. Experimental results suggest that the dynamic search process could effectively search for microblogs, especially for implicitly referred events. Moreover, we show high applicability of our proposed approach to unseen events for which any relevant microblogs were not available in the training phase.

This chapter presents in-depth reviews of search tools for supporting information search. With the brief introduction of cutting-edge search support tools, we describe the key ideas behind the tools and implications for design. We also discuss the limitations of conventional search interfaces to explore directions for future research on search support tools.

Knowledge of entity histories is often necessary for comprehensive understanding and characterization of entities. Yet, the analysis of an entity's history is often most meaningful when carried out in comparison with the histories of other entities. In this paper, we describe a novel task of history-based entity categorization and comparison. Based on a set of entity-related documents which are assumed as an input, we determine latent entity categories whose members share similar histories; hence, we are effectively grouping entities based on the correspondences in their historical developments. Next, we generate comparative timelines for each determined group allowing users to elucidate similarities and differences in the histories of entities. We evaluate our approach on several datasets of different entity types demonstrating its effectiveness against competitive baselines.

Wikipedia contains large amounts of content related to history. It is being used extensively for many knowledge intensive tasks within computer science, digital humanities and related fields. In this paper, we look into Wikipedia articles on historical people for studying link-related temporal features of articles on past people. Our study sheds new light on the characteristics of information about historical people recorded in the English Wikipedia and quantifies user interest in such data. We propose a novel style of analysis in which we use signals derived from the hyperlink structure of Wikipedia as well as from article view logs, and we overlay them over temporal dimension to understand relations between time periods, link structure and article popularity. In the latter part of the paper, we also demonstrate several ways for estimating person importance based on the temporal aspects of the link structure as well as a method for ranking cities using the computed importance scores of their related persons.

General Chairs Marta Mrak, BBC, UK Ebroul Izquierdo, Queen Mary University of London, UK Vladan Velisavljevic, University of Bedfordshire, UK Program Chairs Shuai Wan, Northwestern Polytechnical Uni., China Ce Zhu, Uni. of Elec. Sci. & Tech. of China, China Jian Zhang, Uni. of Technology, Sydney, Australia Jianquan Liu, NEC Corporation, Japan Shiwen Mao, Auburn University, US Wen-Huang Cheng, National Chiao Tung Uni, Taiwan Jiwen Lu, Tsinghua University, China Plenary Chairs Mei-Ling Shyu, University of Miami, US Joao Ascenso, Instituto Superior Tecnico, Portugal Workshop Chairs Noel O'Connor, Dublin City University, Ireland Raouf Hamzaoui, De Montfort University, UK Special Session Chairs Athanasios Mouchtaris, Amazon, US Gene Cheung, York University, Canada Tutorial Chairs Pascal Frossard, EPFL, Switzerland Joao Magalhaes, Uni. Nova de Lisboa, Portugal Panel Chairs Chia-Wen Lin, National Tsing Hua University, Taiwan Nikolaos Boulgouris, Brunel University, UK Industry Chairs Vanessa Testoni, Samsung, Brazil Shenglan Huang, Hulu, China Béatrice Pesquet, Thales, France Demo Chairs Christian Timmerer, Alpen-Adria-Uni. Klagenfurt, Austria Saverio Blasi, BBC, UK Expo Chairs Sebastiaan Van Leuven, Twitter, UK Balu Adsumilli, YouTube, US Finance Chair Qianni Zhang, Queen Mary University of London, UK Publication Chair Eduardo Peixoto, University of Brasilia, Brazil Publicity Chairs Fiona Rivera, BBC, UK Enrico Magli, Politecnico di Torino, Italy Homer Chen, National Taiwan University, Taiwan Joern Ostermann, Leibniz Uni. Hannover, Germany Registration and Local Chairs Tomas Piatrik, Queen Mary University of London, UK Charith Abhayaratne, University of Sheffield, UK Grand Challenge Chairs Dave Bull, University of Bristol, UK Patrick Le Callet, University of Nantes, France European Research Projects Chair Alberto Rabbachin, European Commission, Belgium Award Chair Maria Martini, Kingston University, UK Student Program Chair Hantao Liu, Cardiff University, UK –

Knowledge of entity histories is often necessary for comprehensive understanding and characterization of entities. In this paper we introduce a novel task of history-based entity categorization. Taking a set of entity-related documents as an input we detect latent entity categories whose members share similar histories, effectively, grouping entities based on the similarities of their historical developments. Next, we generate comparative timelines for each generated group allowing users to spot similarities and differences in entity histories. We evaluate our approach on several datasets of different entity types demonstrating its effectiveness against competitive baselines.

This paper proposes methods of finding a ranked list of entities for a given query (e.g."Kennin-ji", "Tenryu-ji", or "Kinkaku-ji" for the query "ancient zen buddhist temples in kyoto") by leveraging different types of modifiers in the query through identifying corresponding properties (e.g.established date and location for the modifiers "ancient" and "kyoto", respectively).While most major search engines provide the entity search functionality that returns a list of entities based on users' queries, entities are neither presented for a wide variety of search queries, nor in the order that users expect.To enhance the effectiveness of entity search, we propose two entity ranking methods.Our first proposed method is a Webbased entity ranking that directly finds relevant entities from Web search results returned in response to the query as a whole, and propagates the estimated relevance to the other entities.The second proposed method is a property-based entity ranking that ranks entities based on properties corresponding to modifiers in the query.To this end, we propose a novel property identification method that identifies a set of relevant properties based on a Support Vector Machine (SVM) using our seven criteria that are effective for different types of modifiers.The experimental results showed that our proposed property identification method could predict more relevant properties than using each of the criteria separately.Moreover, we achieved the best performance for returning a ranked list of relevant entities when using the combination of the Web-based and property-based entity ranking methods.

Categorization is a common solution used for organizing entities like persons or places. For example, there are over 1.13 million categories in Wikipedia which group various types of entities. What is however often lacking is a general, shared information about the entities with a category, for example, information on typical histories of the category entities. We propose in this paper a novel task of automatically creating summaries of shared histories of entities within their categories (e.g., a typical history of a Japanese city). The output summary is in the form of key representative events of entities together with the information on their typical dates. We introduce 4 methods for the aforementioned task and evaluate them on Wikipedia categories containing several types of cities and persons. The summaries we generate can provide information on the common evolution of entities falling into the same category as well as they can be compared with the summaries of related categories for generating contrastive type of knowledge.

This paper proposes methods of finding a ranked list of entities using RDF data for a given sentential query ( e.g. "Cars 3", "Toy Story 4", or "The Incredibles 2" for the query "upcoming animated films pixar") by leveraging different types of modifiers in the query through identifying corresponding properties ( e.g. released and movie type for the modifiers "upcoming" and "animated", respectively). While major search engines provide the entity search functionality that returns a list of entities based on users' queries, entities are neither presented for a wide variety of search queries, nor in the order that users expect. To enhance the efficiency of entity search, we propose two entity ranking methods. Our first proposed method is a Web-based entity ranking that directly finds highly relevant entities from Web search results returned in response to the query as a whole, and propagates the estimated relevance to the other entities. The second proposed method is a property-based entity ranking that ranks entities based on properties corresponding to modifier terms in the query. To this end, we propose a novel method that identifies a set of relevant properties based on the combination of the frequency of property values containing the modifier, co-occurrence of the modifier and property names, and difference in property value distributions of entities in the search results for a query. The experimental results showed that our proposed property identification method could predict more relevant properties than using each criterion separately. Moreover, we achieved the best performance for returning a ranked list of relevant entities when using both of the Web-based and property-based entity ranking methods.

This study proposes a method for retrieving and ranking posts from social network services(SNSs) by specifying and providing feedback on the context of posts. Current search systems for SNS posts cannot handle user intent with regard to the context of posts to be retrieved, mainly owing to the incompleteness of SNS posts, i.e., they do not contain the users' contexts (e.g., situations or preferences) of users posting messages. Hence, we propose a search method that accepts two kinds of queries, namely, content queries and context queries, and that updates these queries based on the user feedback with special attention to the contexts of posts. Our search method considers the whole SNS dataset as a graph and the nodes surrounding each post as its context; to find relevant posts in terms of content and context, our method propagates user feedback via this graph. Our experimental results based on a Twitter test collection revealed that our proposed method showed improved retrieval performance as compared with conventional SNS retrieval and relevance feedback. In addition, we could detect the optimal parameters for feedback propagating.

Wikipedia is the result of collaborative effort aiming to represent human knowledge and to make it accessible to the public. Many Wikipedia articles however lack key metadata information. For example, relatively large number of people described in Wikipedia have no information on their birth and death dates. We propose in this paper to estimate entity's lifetimes using link structure in Wikipedia focusing on person entities. Our approach is based on propagating temporal information over links between Wikipedia articles.