Large language model-based agents are becoming increasingly popular as a low-cost mechanism to provide personalized, conversational advice, and have demonstrated impressive capabilities in relatively simple scenarios, such as movie recommendations. But how do these agents perform in complex high-stakes domains, where domain expertise is essential and mistakes carry substantial risk? This paper investigates the effectiveness of LLM-advisors in the finance domain, focusing on three distinct challenges: (1) eliciting user preferences when users themselves may be unsure of their needs (2) providing personalized guidance for diverse investment preferences, and (3) leveraging advisor personality to build relationships and foster trust. Via a lab-based user study with 64 participants, we show that LLM-advisors often match human advisor performance when eliciting preferences, although they can struggle to resolve conflicting user needs. When providing personalized advice, the LLM was able to positively influence user behavior, but demonstrated clear failure modes. Our results show that accurate preference elicitation is key, otherwise, the LLM-advisor has little impact, or can even direct the investor toward unsuitable assets. More worryingly, users appear insensitive to the quality of advice being given, or worse these can have an inverse relationship. Indeed, users reported a preference for and increased satisfaction as well as emotional trust with LLMs adopting an extroverted persona, even though those agents provided worse advice.

Causality is fundamental in human cognition and has drawn attention in diverse research fields. With growing volumes of textual data, discerning causalities within text data is crucial, and causal text mining plays a pivotal role in extracting meaningful patterns. This study conducts comprehensive evaluations of ChatGPT’s causal text mining capabilities. Firstly, we introduce a benchmark that extends beyond general English datasets, including domain-specific and non-English datasets. We also provide an evaluation framework to ensure fair comparisons between ChatGPT and previous approaches. Finally, our analysis outlines the limitations and future challenges in employing ChatGPT for causal text mining. Specifically, our analysis reveals that ChatGPT serves as a good starting point for various datasets. However, when equipped with a sufficient amount of training data, previous models still surpass ChatGPT’s performance. Additionally, ChatGPT suffers from the tendency to falsely recognize non-causal sequences as causal sequences. These issues become even more pronounced with advanced versions of the model, such as GPT-4. In addition, we highlight the constraints of ChatGPT in handling complex causality types, including both intra/inter-sentential and implicit causality. The model also faces challenges with effectively leveraging in-context learning and domain adaptation. We release our code to support further research and development in this field.

This study focuses on discovering the ripple effects on firms of significant external events such as pandemics, resource price spikes, and natural disasters. We construct a system that presents a chain of other economic events derived from one event by extracting descriptions of causal relationships from a large amount of text data. The objective of this study is to construct a causal chain presentation system that takes into account the diversity of outputs to discover broader economic ripple effects. To accomplish this goal, we develop a new algorithm that uses Maximal Marginal Relevance to represent causal chains. Experiments using Japanese financial statement summaries confirm that this approach produces a greater variety of outputs without sacrificing accuracy. This innovative approach provides a more comprehensive understanding of how external events impact firms and their financial performance.

Recommender systems often struggle with the user cold start problem, which arises when there is a lack of interaction data for new users. This issue is particularly important in financial stock recommendations, as novice investors often lack investment experience and require personalized advice more than experienced users. Behavioral finance offers valuable insights into investor preferences and highlights the impact of psychological factors on investor behavior. In this paper, we present a novel framework that integrates behavioral finance with financial stock recommendations to effectively tackle the user cold start problem. To that end, we first conduct a survey involving 964 Japanese individual investors to gather investment-related behavioral traits while collecting their transaction data in a trading platform. Then, we introduce the Investor Risk-tolerance Aware DropoutNet (IRAD) and show its improved performance over baseline models, demonstrating its effectiveness for stock recommendations in cold start settings. Finally, we offer an example of how incorporating investors’ behavioral traits can result in more interpretable stock recommendations.

With the development of online platforms, people can share and obtain opinions quickly. It also makes individuals' preferences change dynamically and rapidly because they may change their minds when getting convincing opinions from other users. Unlike representative areas of recommendation research such as e-commerce platforms where items' features are fixed, in investment scenarios financial instruments' features such as stock price, also change dynamically over time. To capture these dynamic features and provide a better-personalized recommendation for amateur investors, this study proposes a Personalized Dynamic Recommender System for Investors, PDRSI. The proposed PDRSI considers two investor's personal features: dynamic preferences and historical interests, and two temporal environmental properties: recent discussions on the social media platform and the latest market information. The experimental results support the usefulness of the proposed PDRSI, and the ablation studies show the effect of each module. For reproduction, we follow Twitter's developer policy to share our dataset for future work.

The personalized stock recommendation is a task to recommend suitable stocks for each investor. The personalized recommendations are valuable, especially in investment decision making as the objective of building a portfolio varies by each retail investor. In this paper, we propose a Personalized Stock Recommendation with Investors' Attention and Contextual Information (PSRIC). PSRIC aims to incorporate investors' financial decision-making process into a stock recommendation, and it consists of an investor modeling module and a context module. The investor modeling module models the investor's attention toward various stock information. The context module incorporates stock dynamics and investor profiles. The result shows that the proposed model outperforms the baseline models and verifies the usefulness of both modules in ablation studies.

The gravity models has been studied to analyze interaction between two objects such as trade amount between a pair of countries, human migration between a pair of countries and traffic flow between two cities. Particularly in the international trade, predicting trade amount is instrumental to industry and government in business decision making and determining economic policies. Whereas the gravity models well captures such interaction between objects, the model simplifies the interaction to extract essential relationships or needs handcrafted features to drive the models. Recent studies indicate the connection between graph neural networks (GNNs) and the gravity models in international trade. However, to our best knowledge, hardly any previous studies in the this domain directly predicts trade amount by GNNs. We propose GGAE (Gravity-informed Graph Auto-encoder) and its surrogate model, which is inspired by the gravity model, showing trade amount prediction by the gravity model can be formulated as an edge weight prediction problem in GNNs and solved by GGAE and its surrogate model. Furthermore, we conducted experiments to indicate GGAE with GNNs can improve trade amount prediction compared to the traditional gravity model by considering complex relationships.

We propose a new pre-trained architecture for economics/finance domain adaptation of language models in this paper. Pre-trained language models have become commonplace in a wide range of language processing applications. Because they learn from generic documents such as Wikipedia, many pre-trained language models are not fully adapted to the domain. As a result, there are two approaches: one is to develop a domain-specific pre-trained language model, and the other is to adapt the model learned on general documents to the domain through further pre-training with domain texts. However, no definitive better method has been discovered, and each project is working on it in different ways. As a result, this study focuses on the Japanese language’s economics/financial domains and investigates how pre-trained language models can be better adapted to domain-specific tasks.

The COVID-19 pandemic has affected public behavior in a variety of ways. Concerns about the spread of a hitherto unknown virus drove numerous changes in public behavior, including a greater tendency to self-isolate at home. In this study, we assigned numerical scores to key sentiments expressed in COVID-19-related posts on major social media platform Twitter to measure changes in public sentiment during the pandemic. We also examined the relationship between mobility in various locations around Japan and scores for sentiments such as dislike and fear. Our research provided evidence of a tendency for mobility to decline (i.e. for more people to self-isolate at home) roughly one month after scores for negative public sentiment regarding COVID-19 increased. Mobility is closely connected with a variety of economic activities, mainly in service industries. This suggests that the sentiment in Twitter postings on COVID-19 that we discuss in this study is a leading indicator of changes in mobility (the extent to which people self-isolate at home), demonstrating the effectiveness of Twitter data in forecasting short-term changes in economic activity during the pandemic.

This study demonstrates whether financial text is useful for tactical asset allocation using stocks by using natural language processing to create polarity indexes in financial news. In this study, we performed clustering of the created polarity indexes using the change-point detection algorithm. In addition, we constructed a stock portfolio and rebalanced it at each change point utilizing an optimization algorithm. Consequently, the asset allocation method proposed in this study outperforms the comparative approach. This result suggests that the polarity index helps construct the equity asset allocation method.

Recently, the National Statistical Institutes (NSIs) have started to use new data sources to produce official statistics. These new data sources often referred to as "Big Data" are not directly related to statistical production purposes. An advantage of using Big Data for official statistics is the speed of publication. Surveys designed to produce official statistics are time-consuming. On the other hand, we can obtain such new data sources with almost no time lag as the previous day ’s information is available the next day. However, the new data sources have a problem. They are likely selective with respect to the target population. This selection bias needs to be corrected when using these data sources to produce official statistics. In this study, we implemented Big Data to the Japanese official labor statistics, wage changes of hired career-changing employees. This economic indicator can potentially be indispensable for public policymakers and recruiters in private firms. If the indicator had been published quickly, they could have identified the degree of pressure on wage by the external labor market and could have used the information for their decision-making. The problem is the indicator is not published quickly. The time lag ranges from six to thirteen months. This means the indicator cannot be used to make decisions. To address this problem, we used transaction data of private employment agency and the idea of supervised learning under covariate shift to correct the selection bias. The proposed method can achieve early publication if a certain margin of error is allowed. The range of time lag reduces from a few days to five months, and policymakers and hiring managers could use the indicator for their decision-making.

Stock embedding is a method for vector representation of stocks. There is a growing demand for vector representations of stock, i.e., stock embedding, in wealth management sectors, and the method has been applied to various tasks such as stock price prediction, portfolio optimization, and similar fund identifications. Stock embeddings have the advantage of enabling the quantification of relative relation-ships between stocks, and they can extract useful information from unstructured data such as text and network data. In this study, we propose stock embedding enhanced with textual and network information (SETN) using a domain-adaptive pre-trained transformer-based model to embed textual information and a graph neural network model to grasp network information. We evaluate the performance of our proposed model on related company information extraction tasks. We also demonstrate that stock embeddings obtained from the proposed model perform better in creating thematic funds than those obtained from baseline methods, providing a promising pathway for various applications in the wealth management industry.

In recent years, instead of closing data and analysis skills in-house, there has been much interest in widely releasing data analysis knowledge on the web. A data exchange platform is a type of digital platform that exchanges data between stakeholders, e.g., data owners, users, and analysts. However, the datasets handled on such platforms are independently acquired and stored by the data providers for their own purposes. These datasets are not based on the premise of coordination and combination, and there is currently little information available to discuss the systematic organization and combination of these datasets. In this study, we focus on a metadata, summary information of data, and examine the similarity of data on a data exchange platform using natural language processing. In our experiments, we use the metadata from the data exchange platform Kaggle. To compare the similarity of the data, our method employs word2vec and BERT as vectorize methods and converts data descriptions to vectors. Then, our method measures the distances of each vector by calculating cosine similarities between each vector. From experimental results, we found that Kaggle has the same character as other data exchange platforms. Additionally, the results indicated the usability of the natural language processing-based method for extracting similar data pairs.

This study demonstrates whether analysts’ sentiment toward individual stocks is useful in predicting the macroeconomic index. This can be achieved by using natural language processing to create polarity indexes from analyst reports. In this study, the created polarity indexes were analyzed using the Vector Autoregressive model with various macroeconomic indexes. Consequently, it was confirmed that the polarity indexes do have an impact on indexes such as prices, exchange rates, and government bonds.

With the development of computers and the rise of data exchange, the expectations for innovation by combining data from different industries, and data exchange platforms that handle different types of data are sprouting up. However, the data handled on such platforms have been obtained and are stored independently by data providers with different purposes, and the maintenance of data catalogs and the spread of schemata are not currently insufficient, making it difficult to understand the relationships between the data on such platforms. In this study, in order to derive the relationships among datasets and to discuss the similarity and combinability of the data on the data exchange platform, we analyze and discuss the relationships between datasets on the platform. We focussed on the data outlines and variables using the metadata of the data exchange platform service, D-Ocean, and found that the similarity of data cannot be measured by a single indicator, and that the items to be referred to calculate the similarity depending on the indicator.

Deep neural networks are powerful for text sentiment analysis; however, in the real world, they cannot be used in situations where explanations are required owing to their black-box property. In response, we propose a novel neural network model called sentiment shift neural network (SSNN) that can explain the process of its sentiment analysis prediction in a way that humans find natural and agreeable. The SSNN has the following three interpretable layers: the word-level original sentiment layer, sentiment shift layer, and word-level contextual sentiment layer. Using these layers, the SSNN can explain the process of its document-level sentiment analysis results in a human-like way. Realizing the interpretability of these layers is a crucial problem. To realize this interpretability, we propose a novel learning strategy called joint sentiment propagation (JSP) learning. Using real textual datasets, we experimentally demonstrate that the proposed JSP learning is effective for improving the interpretability of layers in SSNN and that both the predictability and explanation ability of the SSNN are high.

Abstract Although deep neural networks are excellent for text sentiment analysis, their applications in real-world practice are occasionally limited owing to their black-box property. In this study, we propose a novel neural network model called contextual sentiment neural network (CSNN) model that can explain the process of its sentiment analysis prediction in a way that humans find natural and agreeable and can catch up the summary of the contents. The CSNN has the following interpretable layers: the word-level original sentiment layer, word-level sentiment shift layer, word-level global importance layer, word-level contextual sentiment layer, and concept-level contextual sentiment layer. Because of these layers, this network can explain the process of its document-level sentiment analysis results in a human-like way using these layers. Realizing the interpretability of each layer in the CSNN is a crucial problem in the development of this CSNN because the general back-propagation method cannot realize such interpretability. To realize this interpretability, we propose a novel learning strategy called initialization propagation (IP) learning. Using real textual datasets, we experimentally demonstrate that the proposed IP learning is effective for improving the interpretability of each layer in CSNN. We then experimentally demonstrate that the CSNN has both the high predictability and high explanation ability.

Although deep neural networks are excellent for text sentiment analysis, their applications in real-world practice are occasionally limited owing to their black-box property. In response, we propose a novel neural network model called contextual sentiment neural network (CSNN) model that can explain the process of its sentiment analysis prediction in a way that humans find natural and agreeable. The CSNN has the following interpretable layers: the word-level original sentiment layer, word-level sentiment shift layer, word-level local contextual sentiment layer, word-level global importance layer, and word-level global contextual sentiment layer. Because of these layers, this network can explain the process of its document-level sentiment analysis results in a human-like way using these layers. Realizing the interpretability of each layer in the CSNN is a crucial problem in the development of this CSNN because the general back-propagation method cannot realize such interpretability. To realize this interpretability, we propose a novel learning strategy called initialization propagation (IP) learning. Using real textual datasets, we experimentally demonstrate that the proposed IP learning is effective for improving the interpretability of each layer in CSNN. We then experimentally demonstrate that both the predictability and explanation ability of the CSNN are high.

We constructed an evaluation system of the self-impact in a financial market using an artificial market and text-mining technology. Economic trends were first extracted from text data circulating in the real world. Then, the trends were inputted into the market simulation. Our simulation revealed that an operation by intervention could reduce over 70% of rate fluctuation in 1995. By the simulation results, the system was able to help for its user to find the exchange policy which can stabilize the yen-dollar rate