We developed the WISDOM-DX Recommendation System by extracting digital transformation (DX) activities of companies from web information and using generative AI summarization. This system recommends Manabi DX learning contents for DX talent development to learners based on the DX activity status and levels of their companies. The WISDOM-DX Recommendation System first extracts various DX activities for each company from web information and evaluates the activity levels. The extracted activities are then summarized by generative AI to generate a company’s DX summary. The system calculates the similarity between this DX summary and the Manabi DX learning contents. Based on this similarity and the consistency between the DX activity level and the Manabi DX contents level, the system determines the matching degree and recommends learning contents to DX promoters of each company. Using this system, we recommended 10 Manabi DX learning contents from the 732 available in Manabi DX to 63 companies. Sixty-one companies (97

DX is a must for Japanese companies to cope with Japan's 2025 Digital Cliff. To promote DX initiatives, self-assessment is necessary to make their position recognized. However, conventional manual assessments have high costs, hindering their DX promotion. We noticed that progress in corporate DX can be reflected in the amount of related information on the Web, and there is no reported method of utilizing this relation for DX assessment. This paper proposes WISDOM-DX, a system that accumulates information on corporate activities on the Web using a question-answering system, and ranks companies regarding DX initiatives. We compared WISDOM-DX with two baselines Prize and Google. The precision of WISDOM-DX, Prize, and Google were 56.3%, 45.8%, and 22.9%, respectively. The rate of DX-related award winners or certified companies obtained by WISDOM-DX and Google were 91.7% and 64.6%, respectively. The Area Under the Precision-Recall curve (AUPR) of WISDOM-DX, Prize, and Google were 0.540, 0.359, and 0.181, respectively. An opinion survey showed 60.7% positive and 32.1% neutral responses regarding the agreeability of WISDOM-DX's rankings, and 46.4% positive and 39.3% neutral responses regarding the usefulness of WISDOM-DX. These results showed WISDOM-DX's promising performance and the prospect of automating large-scale assessment regarding corporate DX initiatives.

近年，ICT の発達に伴い，災害対応の現場でもこれらを活用した災害対応の効率化が行われている．特にTwitter やLINE に代表されるコミュニケーションツールは被災者一人一人と災害対応機関とのコミュニケーションを変えつつある．本稿では，現在コミュニケーションツールが災害対応の現場で活用されている例やその背景にある技術について紹介するとともに，我々が開発中の防災チャットボットSOCDA の機能と展望についても紹介する．

For effective responses to flood disasters, it is essential to identify affected areas in real time. Recently, social media (e.g. Twitter) have emerged as new sources of disaster-related information in real time. However, concerns still remain regarding the trustworthiness and the amount of information, especially that issued from a site of crisis. This study investigated a total of 109 tweets sampled based on certain criteria during a flood disaster in the Kinu River Basin, Japan in September 2015. We classified them into five categories depending on the main contents: 1) flood inundation, 2) rescue, 3) emotion, 4) river condition, and 5) damage situation. The analysis suggests that the highest proportion (37%) of tweets were related to flood inundation followed by damage (19%) and 32% of them were posted in near real time with photos. We further compared well-positioned tweets with other inundation extent information based on our field investigations and aerial photos. The results showed good agreement between the inundation information from the posted tweets and the expected locations. Some tweets suggested additional inundated areas, not originally identified by the aerial photos. Overall, the study shows the potential use of social media to collect local details about floods.

We demonstrate our large-scale NLP systems: WISDOM X, DISAANA, and D-SUMM. WISDOM X provides numerous possible answers including unpredictable ones to widely diverse natural language questions to provide deep insights about a broad range of issues. DISAANA and D-SUMM enable us to assess the damage caused by large-scale disasters in real time using Twitter as an information source.

After the Great East Japan Earthquake in 2011, an abundance of false rumors were disseminated on Twitter that actually hindered rescue activities. This work presents a method for recognizing the negation of predicates on Twitter to find Japanese tweets that refute false rumors. We assume that the predicate “occur” is negated in the sentence “The guy who tweeted that a nuclear explosion occurred has watched too many SF movies.” The challenge is in the treatment of such complex negation. We have to recognize a wide range of complex negation expressions such as “it is theoretically impossible that...” and “The guy who... watched too many SF movies.” We tackle this problem using a combination of a supervised classifier and clusters of n-grams derived from large un-annotated corpora. The n-gram clusters give us a gain of about 22% in F-score for complex negations.

This work presents a novel language model construction method for speech recognition, utilized with “Ikkyu”, an open-domain speech-based question answering system. Ikkyu accepts relatively short spoken questions concerning a large variety of topics as input through a smartphone, providing the answers retrieved from a large scale Web archive. Our challenge is to construct a language model that can accurately perform speech recognition of open domain questions with smartphones as input devices. We tackle this problem by combining an existing domain adaptation method and distributional word similarity. From 500 seed sentences and a corpus of 600 million Web pages we constructed a language model covering 413,000 words. We achieved an average improvement of 3.25 points in word error rate (WER) over a baseline model constructed from randomly sampled Web sentences.

times of crisis.To tackle this problem we introduce a mechanism by which rescue workers from NPOs or municipalities can register certain questions for situation assessment in advance, so that when a disaster victim posts some urgent requests for food, medicines or other essentials on Twitter or some other BBS, both information sender and information requester are automatically notified of this.We expect that such a mechanism can safeguard the two-way communication between rescue workers and disaster victims, and ultimately lead to a more effective rescue effort.We evaluate the system on a test set of 300 questions and their answers.For 192 questions whose answers are actually included in our system's index, we obtained on average 605.8 answers per question, with 51.9% recall and 60.8% precision.

In this paper, we explore the utility of intra- and inter-sentential causal relations between terms or clauses as evidence for answering why-questions. To the best of our knowledge, this is the first work that uses both intra- and inter-sentential causal relations for why-QA. We also propose a method for assessing the appropriateness of causal relations as answers to a given question using the semantic orientation of excitation proposed by Hashimoto et al. (2012). By applying these ideas to Japanese why-QA, we improved precision by 4.4% against all the questions in our test set over the current state-of-theart system for Japanese why-QA. In addition, unlike the state-of-the-art system, our system could achieve very high precision (83.2%) for 25% of all the questions in the test set by restricting its output to the confident answers only.

The 2011 Great East Japan Earthquake caused a wide range of problems, and as countermeasures, many aid activities were carried out. Many of these problems and aid activities were reported via Twitter. However, most problem reports and corresponding aid messages were not successfully exchanged between victims and local governments or humanitarian organizations, overwhelmed by the vast amount of information. As a result, victims could not receive necessary aid and humanitarian organizations wasted resources on redundant efforts. In this paper, we propose a method for discovering matches between problem reports and aid messages. Our system contributes to problem-solving in a large scale disaster situation by facilitating communication between victims and humanitarian organizations.

Immediately after the 2011 Great East Japan Earthquake, the Internet was flooded by a huge amount of information concerning the damage and problems caused by the earthquake, the tsunami, and the nuclear disaster. Many reports about aid efforts and advice to victims were also transmitted into cyberspace. However, since most people were overwhelmed by the massive amounts of information, they could not make proper decisions, and much confusion was caused. Furthermore, false rumors spread on the Internet and fanned such confusion. We demonstrate NICT’s prototype disaster information analysis system, which was designed to properly organize such a large amount of disaster-related information on social media during future large-scale disasters to help people understand the situation and make correct decisions. We are going to deploy it using a large-scale computer cluster in fiscal year 2014.

We demonstrate our large-scale web information analysis system called WISDOM2013, which consists of several deep semantic analysis systems such as a factoid QA, a non-factoid QA and a sentiment analyzer, and a software platform on which its semantic analysis systems can be applied to a billion-page-scale web archive. The software platform has an extendable architecture, and we are planning to enhance WISDOM2013 in the future by adding more semantic analysis systems and inference mechanisms.

In this paper we propose a two-stage method to acquire contradiction relations between typed lexico-syntactic patterns such as Xdrug prevents Ydisease and Ydisease caused by Xdrug. In the first stage, we train an SVM classifier to detect contradiction pattern pairs in a large web archive by exploiting the excitation polarity (Hashimoto et al., 2012) of the patterns. In the second stage, we enlarge the first stage classifier’s training data with new contradiction pairs obtained by combining the output of the first stage’s classifier and that of an entailment classifier. We acquired this way 750,000 typed Japanese contradiction pattern pairs with an estimated precision of 80%. We plan to release this resource to the NLP community.

This paper presents the results of the user evaluation of spo- ken decision support dialogue systems, which help users select from a set of alternatives. Thus far, we have modeled this deci- sion support dialogue as a partially observable Markov decision process (POMDP), and optimized its dialogue strategy to maxi- mize the value of the user’s decision. In this paper, we present a comparative evaluation of the optimized dialogue strategy with several baseline strategies, and demonstrate that the optimized dialogue strategy that was effective in user simulation experi- ments works well in an evaluation by real users.

This paper describes a novel method of constructing a language model for speech recognition of inputs with a particular style, using a large-scale Web archive. Our target is an open domain voice-activated QA system and our speech recognition module must recognize relatively short, domain independent questions. The central issue is how to prepare a large scale training corpus with low cost, and we tackled this problem by combining an existing domain adaptation method and distributional word similarity. From 500 seed sentences and 600 million Web pages we constructed a language model covering 413,000 words. We achieved an average improvement of 3.25 points in word error rate over a baseline model constructed from randomly sampled Web sentences.

This article presents a user model for user simulation and a system state representation in spoken decision support dialogue systems. When selecting from a group of alternatives, users apply different decision-making criteria with different priorities. At the beginning of the dialogue, however, users often do not have a definite goal or criteria in which they place value, thus they can learn about new features while interacting with the system and accordingly create new criteria. In this article, we present a user model and dialogue state representation that accommodate these patterns by considering the user's knowledge and preferences. To estimate the parameters used in the user model, we implemented a trial sightseeing guidance system, collected dialogue data, and trained a user simulator. Since the user parameters are not observable from the system, the dialogue is modeled as a partially observable Markov decision process (POMDP), and a dialogue state representation was introduced based on the model. We then optimized its dialogue strategy so that users can make better choices. The dialogue strategy is evaluated using a user simulator trained from a large number of dialogues collected using a trial dialogue system.

This paper presents a spoken dialogue framework that helps users in making decisions. Users often do not have a definite goal or criteria for selecting from a list of alternatives. Thus the system has to bridge this knowledge gap and also provide the users with an appropriate alternative together with the reason for this recommendation through dialogue. We present a dialogue state model for such decision making dialogue. To evaluate this model, we implement a trial sightseeing guidance system and collect dialogue data. Then, we optimize the dialogue strategy based on the state model through reinforcement learning with a natural policy gradient approach using a user simulator trained on the collected dialogue corpus.