Analysis of the Financial Information Contained in the Texts of Current Reports: A Deep Learning Approach

An important role in the fundamental analysis is played by the acquisition and analysis of various types of information about the company. Text documents are an increasingly important source of this information. Their accurate and quick analysis is an increasingly important challenge for financial analysts. Research in the area of financial text analysis is based on sentiment analysis. The deep neural networks and the stocks’ cumulative abnormal return are used in this article to analyze the sentiment of financial texts. The proposed approach, unlike those used so far, does not require manual labeling of data or the creation of dictionaries and is free from the subjective assessment of the researcher. Taking into account the broad context of words and their meaning in financial texts, it also eliminates the problem of ambiguity of words in various contexts. The sentiment of financial texts presented in this paper is directly related to the market reaction to the information contained in these texts. For texts belonging to one of the two classes (positive or negative) with the highest probability, the deep learning model gives predictions with a precision of 62% for the positive class and 55% for the negative class. The event study results show that the sentiment calculated under the proposed method can be successfully used to determine the probable direction of the market reaction to the information contained in current reports with a 1 percent significance level. The results can be used in market efficiency research, investment strategy development or support of investment analysts using fundamental analysis.

Download Full-text

Sentiment Analysis of Current Reports Texts with Use of Cumulative Abnormal Return and Deep Neural Network

10.20944/preprints202109.0472.v1 ◽

2021 ◽

Author(s):

Maciej A. Wujec

Keyword(s):

Neural Network ◽

Abnormal Return ◽

Deep Neural Network ◽

Subjective Assessment ◽

Investment Strategy ◽

Strategy Development ◽

Cumulative Abnormal Return ◽

Research Investment ◽

Positive Class ◽

Negative Class

The deep neural network - BERT model (Bidirectional Encoder Representations from Transformers) and the stocks cumulative abnormal return is used in this article to analyze the sentiment of financial texts. The proposed approach, unlike those used so far, does not require the creation of dictionaries, takes into account the broad context of words and their meaning in financial texts, eliminates the problem of ambiguity of words in various contexts, does not require manual labelling of data and is free from the subjective assessment of the researcher. The sentiment of financial texts in the meaning presented in this paper is directly related to the market reaction to the information contained in these texts. For texts belonging to one of the two classes (positive or negative) with the highest probability the BERT model gives the results of predictions with a precision level of 62.38% for the positive class and 55% for the negative class. The results at this level can be used in event study, market efficiency research, investment strategy development or support of investment analysts using fundamental analysis.

Download Full-text

Swapped face detection using deep learning and subjective assessment

EURASIP Journal on Information Security ◽

10.1186/s13635-020-00109-8 ◽

2020 ◽

Vol 2020 (1) ◽

Author(s):

Xinyi Ding ◽

Zohreh Raziei ◽

Eric C. Larson ◽

Eli V. Olinick ◽

Paul Krueger ◽

...

Keyword(s):

Deep Learning ◽

Face Detection ◽

Subjective Assessment

Download Full-text

Knowledge, Attitudes, Behaviors of Women Related to Pregnancy, and Early Childhood Caries Prevention: A Cross-Sectional Pilot Study

Journal of Primary Care & Community Health ◽

10.1177/21501327211013302 ◽

2021 ◽

Vol 12 ◽

pp. 215013272110133

Author(s):

Neel Shimpi ◽

Ingrid Glurich ◽

Catherine Maybury ◽

Min Qi Wang ◽

Kazumasa Hashimoto ◽

...

Keyword(s):

Oral Health ◽

Pilot Study ◽

Health Education ◽

Survey Study ◽

Cross Sectional Survey ◽

Cross Sectional ◽

Health Maintenance ◽

Significance Level ◽

Study Results ◽

And Behaviors

Objective Health education interventions during pregnancy can influence maternal oral health (OH), maternal OH-behaviors and children’s OH. Interventions that can be delivered at anytime and anywhere, for example mobile-health (mHealth) provides an opportunity to address challenges of health education and support activation of women in underserved and rural communities to modify their health behavior. This pilot study was undertaken as a part of a mHealth initiative to determine knowledge, attitudes, and behaviors related to pregnancy and ECC prevention among women attending obstetrics/gynecology (OB/GYN) practices at a large rurally-based clinic. Methods A cross-sectional survey study was voluntarily engaged by women (n = 191) aged 18 to 59 years attending OB/GYN visits, over a 3-week period from 12/2019 to 1/2020. Survey results were analyzed applying descriptive statistics, X2 and Fisher’s Exact tests. The significance level was set at P < .0001 for all analyses. Results Approximately half of respondents were between 18 and 29 years (53%), had a college degree (55%), and 100% reported cell phone use. Whereas 53% and 31%, respectively, indicated that they were “somewhat” or “very” sure of how to prevent ECC in their children, only 9% recognized evidence of early decay and 30% did not know the purpose of fluoride. Overall, only 27% of participants correctly answered the knowledge-based questions. Further, only 57% reported their provider explained things in a way that was easy to understand. Only 24% reported seeing a dentist during their current pregnancy. Conclusions Study results suggested potential gaps in knowledge and behaviors related to ECC prevention and provided baseline data to inform future interventions to improve ECC prevention practices. Notably, majority of participants used their cell phones for making medical/dental appointments and reported using their phones to look up health-related information. This demographic represents a potentially receptive target for mHealth approaches to improve understanding of oral health maintenance during pregnancy and ECC prevention.

Download Full-text

Feasibility of Continual Deep Learning-Based Segmentation for Personalized Adaptive Radiation Therapy in Head and Neck Area

Cancers ◽

10.3390/cancers13040702 ◽

2021 ◽

Vol 13 (4) ◽

pp. 702

Author(s):

Nalee Kim ◽

Jaehee Chun ◽

Jee Suk Chang ◽

Chang Geol Lee ◽

Ki Chang Keum ◽

...

Keyword(s):

Deep Learning ◽

Head And Neck ◽

Organs At Risk ◽

Subjective Assessment ◽

Turing Test ◽

Similar Rate ◽

Dice Similarity Coefficient ◽

Adaptive Planning ◽

Test Set ◽

Using Data

This study investigated the feasibility of deep learning-based segmentation (DLS) and continual training for adaptive radiotherapy (RT) of head and neck (H&N) cancer. One-hundred patients treated with definitive RT were included. Based on 23 organs-at-risk (OARs) manually segmented in initial planning computed tomography (CT), modified FC-DenseNet was trained for DLS: (i) using data obtained from 60 patients, with 20 matched patients in the test set (DLSm); (ii) using data obtained from 60 identical patients with 20 unmatched patients in the test set (DLSu). Manually contoured OARs in adaptive planning CT for independent 20 patients were provided as test sets. Deformable image registration (DIR) was also performed. All 23 OARs were compared using quantitative measurements, and nine OARs were also evaluated via subjective assessment from 26 observers using the Turing test. DLSm achieved better performance than both DLSu and DIR (mean Dice similarity coefficient; 0.83 vs. 0.80 vs. 0.70), mainly for glandular structures, whose volume significantly reduced during RT. Based on subjective measurements, DLS is often perceived as a human (49.2%). Furthermore, DLSm is preferred over DLSu (67.2%) and DIR (96.7%), with a similar rate of required revision to that of manual segmentation (28.0% vs. 29.7%). In conclusion, DLS was effective and preferred over DIR. Additionally, continual DLS training is required for an effective optimization and robustness in personalized adaptive RT.

Download Full-text

Region-of-Interest-Based Cardiac Image Segmentation with Deep Learning

Applied Sciences ◽

10.3390/app11041965 ◽

2021 ◽

Vol 11 (4) ◽

pp. 1965

Author(s):

Raul-Ronald Galea ◽

Laura Diosan ◽

Anca Andreica ◽

Loredana Popa ◽

Simona Manole ◽

...

Keyword(s):

Image Segmentation ◽

Deep Learning ◽

Diagnostic System ◽

Region Of Interest ◽

General Purpose ◽

Medical Image Segmentation ◽

Dice Similarity Coefficient ◽

Learning Methods ◽

Study Results ◽

Whole Heart

Despite the promising results obtained by deep learning methods in the field of medical image segmentation, lack of sufficient data always hinders performance to a certain degree. In this work, we explore the feasibility of applying deep learning methods on a pilot dataset. We present a simple and practical approach to perform segmentation in a 2D, slice-by-slice manner, based on region of interest (ROI) localization, applying an optimized training regime to improve segmentation performance from regions of interest. We start from two popular segmentation networks, the preferred model for medical segmentation, U-Net, and a general-purpose model, DeepLabV3+. Furthermore, we show that ensembling of these two fundamentally different architectures brings constant benefits by testing our approach on two different datasets, the publicly available ACDC challenge, and the imATFIB dataset from our in-house conducted clinical study. Results on the imATFIB dataset show that the proposed approach performs well with the provided training volumes, achieving an average Dice Similarity Coefficient of the whole heart of 89.89% on the validation set. Moreover, our algorithm achieved a mean Dice value of 91.87% on the ACDC validation, being comparable to the second best-performing approach on the challenge. Our approach provides an opportunity to serve as a building block of a computer-aided diagnostic system in a clinical setting.

Download Full-text

Assessing the Role of Pericardial Fat as a Biomarker Connected to Coronary Calcification—A Deep Learning Based Approach Using Fully Automated Body Composition Analysis

Journal of Clinical Medicine ◽

10.3390/jcm10020356 ◽

2021 ◽

Vol 10 (2) ◽

pp. 356

Author(s):

Lennard Kroll ◽

Kai Nassenstein ◽

Markus Jochims ◽

Sven Koitka ◽

Felix Nensa

Keyword(s):

Adipose Tissue ◽

Deep Learning ◽

Coronary Artery ◽

Cardiac Ct ◽

Coronary Calcification ◽

Risk Scores ◽

Agatston Score ◽

Composition Analysis ◽

Significance Level ◽

Tissue Quantification

(1) Background: Epi- and Paracardial Adipose Tissue (EAT, PAT) have been spotlighted as important biomarkers in cardiological assessment in recent years. Since biomarker quantification is an increasingly important method for clinical use, we wanted to examine fully automated EAT and PAT quantification for possible use in cardiovascular risk stratification. (2) Methods: 966 patients with intermediate Framingham risk scores for Coronary Artery Disease referred for coronary calcium scans were included in clinical routine retrospectively. The Coronary Artery Calcium Score (CACS) was extracted and tissue quantification was performed by a deep learning network. (3) Results: The Computed Tomography (CT) segmentations predicted by the network indicated no significant correlation between EAT volume and EAT radiodensity when compared to Agatston score (r = 0.18, r = −0.09). CACS 0 category patients showed significantly lower levels of total EAT and PAT volumes and higher EAT and PAT densities than CACS 1–99 category patients (p < 0.01). Notably, this difference did not reach significance regarding EAT attenuation in male patients. Women older than 50 years, thus more likely to be postmenopausal, were shown to be at higher risk of coronary calcification (p < 0.01, OR = 4.59). CACS 1–99 vs. CACS ≥100 category patients remained below significance level (EAT volume: p = 0.087, EAT attenuation: p = 0.98). (4) Conclusions: Our study proves the feasibility of a fully automated adipose tissue analysis in clinical cardiac CT and confirms in a large clinical cohort that volume and attenuation of EAT and PAT are not correlated with CACS. Broadly available deep learning based rapid and reliable tissue quantification should thus be discussed as a method to assess this biomarker as a supplementary risk predictor in cardiac CT.

Download Full-text

The Effectiveness of Durante Hemodialysis Packed Red Cells (PRC) Transfusion on Hemoglobin Levels

Jurnal Ners dan Kebidanan (Journal of Ners and Midwifery) ◽

10.26699/jnk.v8i1.art.p013-018 ◽

2021 ◽

Vol 8 (1) ◽

pp. 013-018

Author(s):

Pria Wahyu Romadhon Girianto ◽

Mega Wahyu Mulyasari

Keyword(s):

Renal Disease ◽

Chronic Renal Disease ◽

Statistical Tests ◽

Health Condition ◽

P Value ◽

Test Design ◽

Significance Level ◽

Study Results ◽

Post Test ◽

The One

Renal Disease was a chronic disease that the most attacking people in Indonesia. Damage to this vital organ in the human body greatly affected a person's health condition, one of which was anemia. This study aimed to determine the effectiveness of Durante hemodialysis PRC transfusions on hemoglobin levels. The method used was Pre-experimental design, with the One group pre-post test design approach. With a sample of 49 patients who underwent regular hemodialysis at RSUD dr. Iskak Tulungagung. Data were obtained by direct observation. Processed by computerized methods with a statistical t-test, the significance level (α) was 0.05. The study results showed that the hemoglobin levels of the pre-Durante hemodialysis PRC transfusion patients were 4-5 mg/dl (53.06%), and the hemoglobin levels of the post-Durante hemodialysis PRC transfusion patients were 6.1-7 mg/dl (34.69%). The results of statistical tests showed that there was an increase in hemoglobin levels in patients who received Durante hemodialysis PRC transfusion by 1.22 mg/dl because p-value = 0.000 < 0.05 (α) means that there was an effect. It could be concluded that the delivery of Durante hemodialysis PRC transfusion could help increase hemoglobin levels. This finding was very helpful for chronic Renal Disease patients undergoing hemodialysis, who have been using erythropoietin preparations because PRC transfusions were cheaper and more effective when compared to using erythropoietin preparations

Download Full-text

Emergence of New Marketing Distribution Strategies: A Call for A Paradigm Shift

European Journal of Business Management and Research ◽

10.24018/ejbmr.2019.4.6.125 ◽

2019 ◽

Vol 4 (6) ◽

Author(s):

Samuel Affran ◽

Richard Kwabena Asare

Keyword(s):

Consumer Goods ◽

Fast Moving ◽

Service Industry ◽

Sales Performance ◽

Significance Level ◽

Distribution Strategy ◽

Unit Change ◽

Study Results ◽

Fast Moving Consumer Goods ◽

Consumer Goods Industry

The purpose of this research is to empirically formulate new distribution strategies that can service the fast moving consumer goods industry and the service industry as a whole. Inspiration was drawn from the orthodox distribution strategies (intensive, selective, and exclusive) currently used in the service industry. To approve its empirical efficacy the study is set also to determine the impact of these new strategies on sales performance. The study is implemented through a two-stage process of literature review and empirical survey. Evidence was drawn from Ghanaian fast moving consumer goods industry. Sstructured questionnaire was used to gather data from 415 randomly sampled members in the target population. The data obtained were processed using SPSS version 24. Multiple regression analysis was also used to assess its impact on sales performance. The study revealed that inten-electro aggressive strategy with an average mean of 4.02 is the most adopted strategy by the Fast-Moving Consumer Goods Companies followed by selec-electro aggressive distribution strategy(average mean of 3.81) and exclu-electro aggressive distribution strategy(average mean of 3.74) in that order. The study again revealed that there is a positive significant relationship between inten-electro aggressive strategy and sales performance ( = .490, p< 0.05). This newly propounded strategy (inten-electro aggressive strategy) is proven to have a significant impact on sales performance in the fast moving consumer goods industry understudy. Thus, inten-electro aggressive strategy has a moderate positive relationship with sales performance. The statistical implication is that, holding all other variables constant, inten-electro aggressive strategy induces 49.0 % change in sales performance of the fast moving consumer goods firms understudy. Thus, this result proves that a unit change in the effective execution of this strategy will induce 49.1% change in sales performance. In other words when inten-electro aggressive strategy is improved by 1%, sales will be improved by 49.0 percentage change. The significance level of this outcome according to the study results was 0.00 which is less than 0.05 indicating that the variance between the two variables in question was significant. The result again proves that exclu -electro aggressive distribution strategy, impact positively on sales performance. It poses as a reasonable positive inclination to sales performance; thus ( = .595, p<0.05), Thus, holding all other variables (InEADS & SeEADS) constant, Exclu -electro aggressive distribution strategy causes 59.5 % change in sales performance. This result proves that a unit change in the efficacy of exclu -electro aggressive distribution strategy in will induce 59.5% change in the company’s sales. The significance level of this outcome in reference to the study results was 0.000 which is less than the standard value of 0.05 indicating that the variance between the two variables in question was significant. Selec-electro aggressive distribution strategy has a positive significant impact on sales performance thus ( =.532, p< 0.01). It can be said therefore that any improvement made in selec-electro aggressive distribution strategy, will cause sales performance of FMCG companies to increase by 53.2 %.

Download Full-text

PENGARUH INFORMASI LABA TERHADAP REAKSI PASAR DENGAN MANAJEMEN LABA RIIL DAN MANAJEMEN LABA AKRUAL SEBAGAI VARIABEL MODERASI

Jurnal Akuntansi ◽

10.25170/jara.v13i1.485 ◽

2019 ◽

Vol 13 (1) ◽

pp. 1-23

Author(s):

Santi Santi ◽

Kurniawati Kurniawati

Keyword(s):

Earnings Management ◽

Market Reaction ◽

Stock Exchange ◽

Real Earnings Management ◽

Manufacturing Companies ◽

Statistical Tool ◽

Significance Level ◽

Moderating Variables ◽

Earnings Information ◽

Final Sample

This study aims to investigate the effect of earnings information on market reaction with accrual and real earnings management as the moderating variables. The sample of this study is manufacturing companies listed in the Indonesia Stock Exchange in 2012-2015. Samples are collected by purposive sampling and resulted in 58 companies as the final sample. Data were analyzed using Moderated Regression Analysis (MRA) for testing hypothesis with significance level 5%. The statistical tool used is SPSS 22. The results of this study shown that market reacts positively significant toward earnings management and real earnings management in aggregate weaken the effect of earnings information toward market reaction. Real earnings management through discretionary expenses strengthen the effect of earnings information toward market reaction. Meanwhile, real earnings management through sales manipulation and overproduction, and accrual earnings management do not moderate the effect of earnings information toward market reaction.

Download Full-text

ViraMiner: Deep Learning on Raw DNA Sequences for Identifying Viral Genomes in Human Samples

10.1101/602656 ◽

2019 ◽

Cited By ~ 4

Author(s):

Ardi Tampuu ◽

Zurab Bzhalava ◽

Joakim Dillner ◽

Raul Vicente

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Dna Sequences ◽

Recommendation System ◽

Clinical Importance ◽

Training Dataset ◽

Viral Genomes ◽

Human Samples ◽

Types Of Information ◽

Improved Accuracy

ABSTRACTDespite its clinical importance, detection of highly divergent or yet unknown viruses is a major challenge. When human samples are sequenced, conventional alignments classify many assembled contigs as “unknown” since many of the sequences are not similar to known genomes. In this work, we developed ViraMiner, a deep learning-based method to identify viruses in various human biospecimens. ViraMiner contains two branches of Convolutional Neural Networks designed to detect both patterns and pattern-frequencies on raw metagenomics contigs. The training dataset included sequences obtained from 19 metagenomic experiments which were analyzed and labeled by BLAST. The model achieves significantly improved accuracy compared to other machine learning methods for viral genome classification. Using 300 bp contigs ViraMiner achieves 0.923 area under the ROC curve. To our knowledge, this is the first machine learning methodology that can detect the presence of viral sequences among raw metagenomic contigs from diverse human samples. We suggest that the proposed model captures different types of information of genome composition, and can be used as a recommendation system to further investigate sequences labeled as “unknown” by conventional alignment methods. Exploring these highly-divergent viruses, in turn, can enhance our knowledge of infectious causes of diseases.

Download Full-text