arXiv subjects

Masato Kikuchi

Publications and source records attributed to Masato Kikuchi.

At least 19 recordsLinked to original sources

arXiv2026

Customer Analysis and Text Generation for Small Retail Stores Using LLM-Generated Marketing Presence

Point of purchase (POP) materials can be created to assist non-experts by combining large language models (LLMs) with human insight. Persuasive POP texts require both customer understanding and expressive writing skills. However, LLM-generated texts often lack creative diversity, while human users may have limited experience in marketing and content creation. To address these complementary limitations, we propose a prototype system for small retail stores that enhances POP creation through human-AI collaboration. The system supports users in understanding target customers, generating draft POP texts, refining expressions, and evaluating candidates through simulated personas. Our experimental results show that this process significantly improves text quality: the average evaluation score increased by 2.37 points on a -3 to +3 scale compared to that created without system support.

cs.HC↗

arXiv2026

On-Demand Instructional Material Providing Agent Based on MLLM for Tutoring Support

Effective instruction in tutoring requires promptly providing instructional materials that match the needs of each student (e.g., in response to questions). In this study, we introduce an agent that automatically delivers supplementary materials on demand during one-on-one tutoring sessions. Our agent uses a multimodal large language model to analyze spoken dialogue between the instructor and the student, automatically generate search queries, and retrieve relevant Web images. Evaluation experiments demonstrate that our agent reduces the average image retrieval time by 44.4 s compared to cases without support and successfully provides images of acceptable quality in 85.7% of trials. These results indicate that our agent effectively supports instructors during tutoring sessions.

cs.HC↗

arXiv2026

Script-to-Slide Grounding: Grounding Script Sentences to Slide Objects for Automatic Instructional Video Generation

While slide-based videos augmented with visual effects are widely utilized in education and research presentations, the video editing process -- particularly applying visual effects to ground spoken content to slide objects -- remains highly labor-intensive. This study aims to develop a system that automatically generates such instructional videos from slides and corresponding scripts. As a foundational step, this paper proposes and formulates Script-to-Slide Grounding (S2SG), defined as the task of grounding script sentences to their corresponding slide objects. Furthermore, as an initial step, we propose ``Text-S2SG,'' a method that utilizes a large language model (LLM) to perform this grounding task for text objects. Our experiments demonstrate that the proposed method achieves high performance (F1-score: 0.924). The contribution of this work is the formalization of a previously implicit slide-based video editing process into a computable task, thereby paving the way for its automation.

cs.CV↗

arXiv2025

CEFR-Annotated WordNet: LLM-Based Proficiency-Guided Semantic Database for Language Learning

Although WordNet is a valuable resource because of its structured semantic networks and extensive vocabulary, its fine-grained sense distinctions can be challenging for second-language learners. To address this issue, we developed a version of WordNet annotated with the Common European Framework of Reference for Languages (CEFR), integrating its semantic networks with language-proficiency levels. We automated this process using a large language model to measure the semantic similarity between sense definitions in WordNet and entries in the English Vocabulary Profile Online. To validate our approach, we constructed a large-scale corpus containing both sense and CEFR-level information from the annotated WordNet and used it to develop contextual lexical classifiers. Our experiments demonstrate that models fine-tuned on this corpus perform comparably to those fine-tuned on gold-standard annotations. Furthermore, by combining this corpus with the gold-standard data, we developed a practical classifier that achieves a Macro-F1 score of 0.81. This result provides indirect evidence that the transferred labels are largely consistent with the gold-standard levels. The annotated WordNet, corpus, and classifiers are publicly available to help bridge the gap between natural language processing and language education, thereby facilitating more effective and efficient language learning.

cs.CL↗

arXiv2024

Coarse-Grained Sense Inventories Based on Semantic Matching between English Dictionaries

WordNet is one of the largest handcrafted concept dictionaries visualizing word connections through semantic relationships. It is widely used as a word sense inventory in natural language processing tasks. However, WordNet's fine-grained senses have been criticized for limiting its usability. In this paper, we semantically match sense definitions from Cambridge dictionaries and WordNet and develop new coarse-grained sense inventories. We verify the effectiveness of our inventories by comparing their semantic coherences with that of Coarse Sense Inventory. The advantages of the proposed inventories include their low dependency on large-scale resources, better aggregation of closely related senses, CEFR-level assignments, and ease of expansion and improvement.

cs.CL↗

arXiv2023

Lyricist-Singer Entropy Affects Lyric-Lyricist Classification Performance

Although lyrics represent an essential component of music, few music information processing studies have been conducted on the characteristics of lyricists. Because these characteristics may be valuable for musical applications, such as recommendations, they warrant further study. We considered a potential method that extracts features representing the characteristics of lyricists from lyrics. Because these features must be identified prior to extraction, we focused on lyricists with easily identifiable features. We believe that it is desirable for singers to perform unique songs that share certain characteristics specific to the singer. Accordingly, we hypothesized that lyricists account for the unique characteristics of the singers they write lyrics for. In other words, lyric-lyricist classification performance or the ease of capturing the features of a lyricist from the lyrics may depend on the variety of singers. In this study, we observed a relationship between lyricist-singer entropy or the variety of singers associated with a single lyricist and lyric-lyricist classification performance. As an example, the lyricist-singer entropy is minimal when the lyricist writes lyrics for only one singer. In our experiments, we grouped lyricists among five groups in terms of lyricist-singer entropy and assessed the lyric-lyricist classification performance within each group. Consequently, the best F1 score was obtained for the group with the lowest lyricist-singer entropy. Our results suggest that further analyses of the features contributing to lyric-lyricist classification performance on the lowest lyricist-singer entropy group may improve the feature extraction task for lyricists.

cs.SD↗

arXiv2022

Improving Multi-class Classifier Using Likelihood Ratio Estimation with Regularization

The universal-set naive Bayes classifier (UNB)~\cite{Komiya:13}, defined using likelihood ratios (LRs), was proposed to address imbalanced classification problems. However, the LR estimator used in the UNB overestimates LRs for low-frequency data, degrading the classification performance. Our previous study~\cite{Kikuchi:19} proposed an effective LR estimator even for low-frequency data. This estimator uses regularization to suppress the overestimation, but we did not consider imbalanced data. In this paper, we integrated the estimator with the UNB. Our experiments with imbalanced data showed that our proposed classifier effectively adjusts the classification scores according to the class balance using regularization parameters and improves the classification performance.

cs.LG↗

arXiv2022

Conservative Likelihood Ratio Estimator for Infrequent Data Slightly above a Frequency Threshold

A naive likelihood ratio (LR) estimation using the observed frequencies of events can overestimate LRs for infrequent data. One approach to avoid this problem is to use a frequency threshold and set the estimates to zero for frequencies below the threshold. This approach eliminates the computation of some estimates, thereby making practical tasks using LRs more efficient. However, it still overestimates LRs for low frequencies near the threshold. This study proposes a conservative estimator for low frequencies, slightly above the threshold. Our experiment used LRs to predict the occurrence contexts of named entities from a corpus. The experimental results demonstrate that our estimator improves the prediction accuracy while maintaining efficiency in the context prediction task.

stat.ML↗

arXiv2022

Developing a Component Comment Extractor from Product Reviews on E-Commerce Sites

Consumers often read product reviews to inform their buying decision, as some consumers want to know a specific component of a product. However, because typical sentences on product reviews contain various details, users must identify sentences about components they want to know amongst the many reviews. Therefore, we aimed to develop a system that identifies and collects component and aspect information of products in sentences. Our BERT-based classifiers assign labels referring to components and aspects to sentences in reviews and extract sentences with comments on specific components and aspects. We determined proper labels based for the words identified through pattern matching from product reviews to create the training data. Because we could not use the words as labels, we carefully created labels covering the meanings of the words. However, the training data was imbalanced on component and aspect pairs. We introduced a data augmentation method using WordNet to reduce the bias. Our evaluation demonstrates that the system can determine labels for road bikes using pattern matching, covering more than 88\% of the indicators of components and aspects on e-commerce sites. Moreover, our data augmentation method can improve the-F1-measure on insufficient data from 0.66 to 0.76.

cs.CL↗

arXiv2021

Online Classroom Evaluation System Based on Multi-Reaction Estimation

Compared with traditional face-to-face teaching, online learning is more convenient. However, during online classes, it is more difficult for teachers to observe all student reactions at the same time. Our system is designed to help teachers to adjust the speed of their lessons by detecting student reactions. In this study, we estimate student head pose, hand poses, and expressions through the camera, all these poses will be used as criteria for judging the student participation. After estimating, we proposed a method to evaluate classroom participation based on student head pose, hand poses, and facial expression recognition. The estimated result divides the class quality into positive, neutral, and negative, then, under the help of the system, teachers can rearrange the content of the class.

cs.CY↗

arXiv2021

Product Information Browsing Support System Using Analytic Hierarchy Process

Large-scale e-commerce sites can collect and analyze a large number of user preferences and behaviors, and thus can recommend highly trusted products to users. However, it is very difficult for individuals or non-corporate groups to obtain large-scale user data. Therefore, we consider whether knowledge of the decision-making domain can be used to obtain user preferences and combine it with content-based filtering to design an information retrieval system. This study describes the process of building a product information browsing support system with high satisfaction based on product similarity and multiple other perspectives about products on the Internet. We present the architecture of the proposed system and explain the working principle of its constituent modules. Finally, we demonstrate the effectiveness of the proposed system through an evaluation experiment and a questionnaire.

cs.IR↗

arXiv2021

Matching Social Issues to Technologies for Civic Tech by Association Rule Mining using Weighted Casual Confidence

More than 80 civic tech communities in Japan are developing information technology (IT) systems to solve their regional issues. Collaboration among such communities across different regions assists in solving their problems because some groups have limited IT knowledge and experience for this purpose. Our objective is to realize a civic tech matchmaking system to assist such communities in finding better partners with IT experience in their issues. In this study, as the first step toward collaboration, we acquire relevant social issues and information technologies by association rule mining. To meet our challenge, we supply a questionnaire to members of civic tech communities and obtain answers on their faced issues and their available technologies. Subsequently, we match the relevant issues and technologies from the answers. However, most of the issues and technologies in this questionnaire data are infrequent, and there is a significant bias in their occurrence. Here, it is difficult to extract truly relevant issues--technologies combinations with existing interestingness measures. Therefore, we introduce a new measure called weighted casual confidence, and show that our measure is effective for mining relevant issues--technologies pairs.

cs.CY↗

arXiv2021

Feature Selective Likelihood Ratio Estimator for Low- and Zero-frequency N-grams

In natural language processing (NLP), the likelihood ratios (LRs) of N-grams are often estimated from the frequency information. However, a corpus contains only a fraction of the possible N-grams, and most of them occur infrequently. Hence, we desire an LR estimator for low- and zero-frequency N-grams. One way to achieve this is to decompose the N-grams into discrete values, such as letters and words, and take the product of the LRs for the values. However, because this method deals with a large number of discrete values, the running time and memory usage for estimation are problematic. Moreover, use of unnecessary discrete values causes deterioration of the estimation accuracy. Therefore, this paper proposes combining the aforementioned method with the feature selection method used in document classification, and shows that our estimator provides effective and efficient estimation results for low- and zero-frequency N-grams.

cs.CL↗

arXiv2021

Developing a Lecture Video Recording System Using Augmented Reality

Assistive technology is a prerequisite for making a high-quality lecture video. It is therefore imperative to edit the lecture video after recording. In this study, we aim to reduce the cumbersome task of lecture video editing by developing a system that enables the addition of visual effects in the video while recording. In particular, we use augmented reality (AR) technology to digitize and display in real-time lecture materials, assistant agents, and other recording contents used by the lecturer. Our system realizes such a mechanism as a lecture recording environment. In addition, our system based on AR technology can support the work of the lecturer, which is difficult to do by oneself while conducting the lecture, using the information of the lecturer's position and the progress of the lecture. We evaluated the system functionality and performance, and verified the system's correct behavior. If the burden of making lecture videos can be reduced, the lecturer will be able to devote more time to improving the quality of lecture contents, which is expected to contribute to the improvement of lectures.

cs.HC↗

arXiv2021

Development of an Extractive Title Generation System Using Titles of Papers of Top Conferences for Intermediate English Students

The formulation of good academic paper titles in English is challenging for intermediate English authors (particularly students). This is because such authors are not aware of the type of titles that are generally in use. We aim to realize a support system for formulating more effective English titles for intermediate English and beginner authors. This study develops an extractive title generation system that formulates titles from keywords extracted from an abstract. Moreover, we realize a title evaluation model that can evaluate the appropriateness of paper titles. We train the model with titles of top-conference papers by using BERT. This paper describes the training data, implementation, and experimental results. The results show that our evaluation model can identify top-conference titles more effectively than intermediate English and beginner students.

cs.CL↗

arXiv2021

Unified Likelihood Ratio Estimation for High- to Zero-frequency N-grams

Likelihood ratios (LRs), which are commonly used for probabilistic data processing, are often estimated based on the frequency counts of individual elements obtained from samples. In natural language processing, an element can be a continuous sequence of $N$ items, called an $N$-gram, in which each item is a word, letter, etc. In this paper, we attempt to estimate LRs based on $N$-gram frequency information. A naive estimation approach that uses only $N$-gram frequencies is sensitive to low-frequency (rare) $N$-grams and not applicable to zero-frequency (unobserved) $N$-grams; these are known as the low- and zero-frequency problems, respectively. To address these problems, we propose a method for decomposing $N$-grams into item units and then applying their frequencies along with the original $N$-gram frequencies. Our method can obtain the estimates of unobserved $N$-grams by using the unit frequencies. Although using only unit frequencies ignores dependencies between items, our method takes advantage of the fact that certain items often co-occur in practice and therefore maintains their dependencies by using the relevant $N$-gram frequencies. We also introduce a regularization to achieve robust estimation for rare $N$-grams. Our experimental results demonstrate that our method is effective at solving both problems and can effectively control dependencies.

cs.CL↗

arXiv2019

Journal Name Extraction from Japanese Scientific News Articles

In Japanese scientific news articles, although the research results are described clearly, the article's sources tend to be uncited. This makes it difficult for readers to know the details of the research. In this paper, we address the task of extracting journal names from Japanese scientific news articles. We hypothesize that a journal name is likely to occur in a specific context. To support the hypothesis, we construct a character-based method and extract journal names using this method. This method only uses the left and right context features of journal names. The results of the journal name extractions suggest that the distribution hypothesis plays an important role in identifying the journal names.

cs.CL↗

arXiv2017

Finding Association Rules by Direct Estimation of Likelihood Ratios

In this paper, we propose a cost function that corresponds to the mean square errors between estimated values and true values of conditional probability in a discrete distribution. We then obtain the values that minimize the cost function. This minimization approach can be regarded as the direct estimation of likelihood ratios because the estimation of conditional probability can be regarded as the estimation of likelihood ratio by the definition of conditional probability. When we use the estimated value as the strength of association rules for data mining, we find that it outperforms a well-used method called Apriori.

stat.AP↗