scholarly journals Beyond Polarity: Interpretable Financial Sentiment Analysis with Hierarchical Query-driven Attention

Author(s):  
Ling Luo ◽  
Xiang Ao ◽  
Feiyang Pan ◽  
Jin Wang ◽  
Tong Zhao ◽  
...  

Sentiment analysis has played a significant role in financial applications in recent years. The informational and emotive aspects of news texts may affect the prices, volatilities, volume of trades, and even potential risks of financial subjects. Previous studies in this field mainly focused on identifying polarity~(e.g. positive or negative). However, as financial decisions broadly require justifications, only plausible polarity cannot provide enough evidence during the decision making processes of humanity. Hence an explainable solution is in urgent demand. In this paper, we present an interpretable neural net framework for financial sentiment analysis. First, we design a hierarchical model to learn the representation of a document from multiple granularities. In addition, we propose a query-driven attention mechanism to satisfy the unique characteristics of financial documents. With the domain specified questions provided by the financial analysts, we can discover different spotlights for queries from different aspects. We conduct extensive experiments on a real-world dataset. The results demonstrate that our framework can learn better representation of the document and unearth meaningful clues on replying different users? preferences. It also outperforms the state-of-the-art methods on sentiment prediction of financial documents. 

2012 ◽  
Vol 2012 ◽  
pp. 1-24 ◽  
Author(s):  
Mona Riabacke ◽  
Mats Danielson ◽  
Love Ekenberg

Comparatively few of the vast amounts of decision analytical methods suggested have been widely spread in actual practice. Some approaches have nevertheless been more successful in this respect than others. Quantitative decision making has moved from the study of decision theory founded on a single criterion towards decision support for more realistic decision-making situations with multiple, often conflicting, criteria. Furthermore, the identified gap between normative and descriptive theories seems to suggest a shift to more prescriptive approaches. However, when decision analysis applications are used to aid prescriptive decision-making processes, additional demands are put on these applications to adapt to the users and the context. In particular, the issue of weight elicitation is crucial. There are several techniques for deriving criteria weights from preference statements. This is a cognitively demanding task, subject to different biases, and the elicited values can be heavily dependent on the method of assessment. There have been a number of methods suggested for assessing criteria weights, but these methods have properties which impact their applicability in practice. This paper provides a survey of state-of-the-art weight elicitation methods in a prescriptive setting.


2019 ◽  
Vol 58 (11) ◽  
pp. 2455-2471
Author(s):  
Teresa León ◽  
Vicente Liern ◽  
Blanca Pérez-Gladish

Purpose In recent years there has been a significant acceleration in the market growth of social impact investing. Policy makers, regulatory bodies and national decision-makers should base their decision-making processes on multiple criteria. These criteria are, by nature, imprecise, ambiguous and uncertain. The purpose of this paper is to provide decision-makers with a mathematical tool which aids them in their decision-making processes identifying the degree of appropriateness of less developed countries in terms of potential success of investment in vaccination campaigns. Design/methodology/approach In this work, the authors have developed a decision-making tool within the framework of multiple criteria decision making and Fuzzy Logic, which aims to aid decision-makers for vaccinations campaigns in less developed countries. In particular, the authors have proposed a Technique for Order Preference by Similarity to Ideal Solution-based method which is able to work in fuzzy environment in order to assess and rank countries based on their fuzzy degree of appropriateness for impact investing in vaccines. Findings The impact investing market provides capital from private sources to address many pressing global challenges such as access to basic services as health. Governments have, therefore, an essential role in supporting the development of this market by improving the risk/return profile of investments through access to credit facilities, tax credits or subsidies or defining the regulation of the supply of investments, provision of technical assistance to investing private companies and co-financing. The proposed framework permits funding decision making taking into account the degree of preparedness and adequacy for impact investing in vaccines of the selected countries. Research limitations/implications Impact investing can play a key role in the reduction of immunization gap offering suitable strategies for both, governments and private investors for the achievement of United Nations Sustainable Development Goals (SDGs). However, in order to make good financial decisions managers should take into account not only health, income, education and other social criteria but also the degree of basic preparedness of the countries in order to ensure the success of the immunization campaigns which means taking into account availability of basic infrastructures, access to electricity, political stability among other criteria. Practical implications However, in order to make good financial decisions managers should take into account not only health, income, education and other social criteria but also the degree of basic preparedness of the countries in order to ensure the success of the immunization campaigns which means taking into account availability of basic infrastructures, access to electricity, political stability among other criteria. Originality/value The proposed model will allow public and private decision makers to make better investment decisions in terms of effectiveness as the provided ranking of countries candidates for the investments is more realistic and takes into account more decision dimensions.


Author(s):  
Guillermo Mateu ◽  
Lucas Monzani ◽  
Roger Muñoz Navarro

In this article, we explain the important role neuroscience plays in economic and financial environments. Hence, we present neuroeconomics as a way to describe how decision-making processes affect brain activity, focusing especially on the importance of economic and financial decisions. We answer some questions regarding the role of emotions in finance, the psychological factors present in financial markets, and how neuropsychological stimuli affect our economic decisions. We conclude by citing the main research in the area of neuroscience in financial decision-making processes, and highlight further research projects in these areas.


2021 ◽  
Vol 11 (21) ◽  
pp. 10397
Author(s):  
Barry Ezell ◽  
Christopher J. Lynch ◽  
Patrick T. Hester

Computational models and simulations often involve representations of decision-making processes. Numerous methods exist for representing decision-making at varied resolution levels based on the objectives of the simulation and the desired level of fidelity for validation. Decision making relies on the type of decision and the criteria that is appropriate for making the decision; therefore, decision makers can reach unique decisions that meet their own needs given the same information. Accounting for personalized weighting scales can help to reflect a more realistic state for a modeled system. To this end, this article reviews and summarizes eight multi-criteria decision analysis (MCDA) techniques that serve as options for reaching unique decisions based on personally and individually ranked criteria. These techniques are organized into a taxonomy of ratio assignment and approximate techniques, and the strengths and limitations of each are explored. We compare these techniques potential uses across the Agent-Based Modeling (ABM), System Dynamics (SD), and Discrete Event Simulation (DES) modeling paradigms to inform current researchers, students, and practitioners on the state-of-the-art and to enable new researchers to utilize methods for modeling multi-criteria decisions.


2019 ◽  
Vol 5 (1) ◽  
pp. 1-5
Author(s):  
Adam RADOMYSKI ◽  
Krzysztof CUR

This article outlines the results of studies concerning supporting the decision-making process in air defense with the use of state-of-the art computer simulator. The simulator is intended to simulate air force operations and air defense in the air, and the simulated facilities are supposed to reflect real and hypothetical facilities. It allows us to conduct experiments with the use of models showing particular fragments of reality, which reduce information entropy characteristic of contemporary decision-making situations in air defense.


Author(s):  
Elide Garbani-Nerini ◽  
Elena Marchiori ◽  
Lorenzo Cantoni

AbstractThis research investigates the state of the art among Switzerland (CH)’s and Liechtenstein (FL)’s destinations, intended here as Destination Marketing Organizations (DMOs), when it comes to their relationship with data: what data are collected, how they are stored, analyzed and what impact they have on the destination. This study aims at bringing insights into smart tourism studies as a key aspect of the debate is how DMOs deal with data. Based on a survey performed with CH’s and FL’s DMOs and related stakeholders, results suggested that there are common conceptual nodes shared by practitioners when it comes to defining smart destinations. However, when it comes to data-related practices (data collection, storage, analysis and sharing) DMOs have very different processes in place. There are organizations that collect but do not extensively analyze data, while others are still not so keen on sharing their data with the whole destination ecosystem. Furthermore, organizations’ decision-making processes appear to be based to some extent on data, especially when it comes to (digital) marketing initiatives and campaigns, although behaviors are quite different also in this area. Destination managers might benefit from this paper as the study shows how to investigate data-related practices of an organization. This type of analysis could allow an assessment of the situation and an understanding of the direction in which the organization might move forward.


Author(s):  
Karl Gustafson

Enlarging upon experiments and analysis that I did jointly some years ago, in which artificial (symbolic, neural-net and pattern) learning and generalization were compared with that of humans, I will emphasize the role of imagination (or lack thereof) in artificial, human and quantum cognition and decision-making processes. Then I will look in more detail at some of the ‘engineering details’ of its implementation (or lack thereof) in each of these settings. In other words, the question posed is: What is actually happening? For example, we previously found that humans overwhelmingly seek, create or imagine context in order to provide meaning when presented with abstract, apparently incomplete, contradictory or otherwise untenable decision-making situations. Humans are intolerant of contradiction and will greatly simplify to avoid it. They can partially correlate but do not average. Human learning is not Boolean. These and other human reasoning properties will then be taken to critique how well artificial intelligence methods and quantum mechanical modelling might compete with them in decision-making tasks within psychology and economics.


Author(s):  
Jennifer M. Roche ◽  
Arkady Zgonnikov ◽  
Laura M. Morett

Purpose The purpose of the current study was to evaluate the social and cognitive underpinnings of miscommunication during an interactive listening task. Method An eye and computer mouse–tracking visual-world paradigm was used to investigate how a listener's cognitive effort (local and global) and decision-making processes were affected by a speaker's use of ambiguity that led to a miscommunication. Results Experiments 1 and 2 found that an environmental cue that made a miscommunication more or less salient impacted listener language processing effort (eye-tracking). Experiment 2 also indicated that listeners may develop different processing heuristics dependent upon the speaker's use of ambiguity that led to a miscommunication, exerting a significant impact on cognition and decision making. We also found that perspective-taking effort and decision-making complexity metrics (computer mouse tracking) predict language processing effort, indicating that instances of miscommunication produced cognitive consequences of indecision, thinking, and cognitive pull. Conclusion Together, these results indicate that listeners behave both reciprocally and adaptively when miscommunications occur, but the way they respond is largely dependent upon the type of ambiguity and how often it is produced by the speaker.


Sign in / Sign up

Export Citation Format

Share Document