Natural Language Processing Analysis Research Articles

Emotional tone in clinical high risk for psychosis: novel insights from a natural language analysis approach

BackgroundIndividuals at clinical high risk (CHR) for psychosis experience subtle emotional disturbances that are traditionally difficult to assess, but natural language processing (NLP) methods may provide novel insight into these symptoms. We predicted that CHR individuals would express more negative emotionality and less emotional language when compared to controls. We also examined associations with symptomatology.MethodsParticipants included 49 CHR individuals and 42 healthy controls who completed a semi-structured narrative interview. Interview transcripts were analyzed using Linguistic Inquiry and Word Count (LIWC) to assess the emotional tone of the language (tone -the ratio of negative to positive language) and count positive/negative words used. Participants also completed clinical symptom assessments to determine CHR status and characterize symptoms (i.e., positive and negative symptom domains).ResultsThe CHR group had more negative emotional tone compared to healthy controls (t=2.676, p=.009), which related to more severe positive symptoms (r2=.323, p=.013). The percentages of positive and negative words did not differ between groups (p’s>.05).ConclusionsLanguage analyses provided accessible, ecologically valid insight into affective dysfunction and psychosis risk symptoms. Natural language processing analyses unmasked differences in language for CHR that captured language tendencies that were more nuanced than the words that are chosen.

Emerging Trends of Self-Harm Using Sodium Nitrite in an Online Suicide Community: Observational Study Using Natural Language Processing Analysis.

There is growing concern around the use of sodium nitrite (SN) as an emerging means of suicide, particularly among younger people. Given the limited information on the topic from traditional public health surveillance sources, we studied posts made to an online suicide discussion forum, "Sanctioned Suicide," which is a primary source of information on the use and procurement of SN. This study aims to determine the trends in SN purchase and use, as obtained via data mining from subscriber posts on the forum. We also aim to determine the substances and topics commonly co-occurring with SN, as well as the geographical distribution of users and sources of SN. We collected all publicly available from the site's inception in March 2018 to October 2022. Using data-driven methods, including natural language processing and machine learning, we analyzed the trends in SN mentions over time, including the locations of SN consumers and the sources from which SN is procured. We developed a transformer-based source and location classifier to determine the geographical distribution of the sources of SN. Posts pertaining to SN show a rise in popularity, and there were statistically significant correlations between real-life use of SN and suicidal intent when compared to data from the Centers for Disease Control and Prevention (CDC) Wide-Ranging Online Data for Epidemiologic Research (⍴=0.727; P<.001) and the National Poison Data System (⍴=0.866; P=.001). We observed frequent co-mentions of antiemetics, benzodiazepines, and acid regulators with SN. Our proposed machine learning-based source and location classifier can detect potential sources of SN with an accuracy of 72.92% and showed consumption in the United States and elsewhere. Vital information about SN and other emerging mechanisms of suicide can be obtained from online forums.

Comprehensive analysis of natural language processing

Natural Language Processing (NLP) is a fascinating field of study that teaches computers to understand and use human language. This means that computers can read, write, and even translate text just like humans. NLP has many practical uses, such as categorizing text, identifying the tone of language, recognizing names in text, translating languages, and answering questions. NLP has come a long way since it was first developed. In the past, it relied on strict rules to understand language, but now it uses advanced techniques like machine learning and deep learning to understand text. However, there are still some challenges in NLP, such as understanding the meaning of words in context and considering cultural differences. Despite these challenges, NLP is being used in many different areas, from healthcare and finance to education and customer service. NLP is transforming the way humans interact with computers and is making it easier to extract important information from large amounts of text.

Healthcare Virtual Assistant using Generative AI

Abstract: Documentation is a clumsy task for the front- desk staff. They've to fill in patient data, schedule appointments, and attend to patient queries. Even healthcare providers must enter EHR data, which takes a lot of time, and they end up spending lesser time with their patients. In the evolving landscape of healthcare administration, the burden of administrative tasks continues to challenge healthcare professionals and institutions. However, with generative AI, doctors can produce clones of patient data and automate form- filling tasks. It can also be integrated with EHR for documentation work. This project introduces a result — a generative artificial intelligence (AI) system designed to ease administrative burdens and enhance operating efficiency. By employing the power of AI, this web- grounded platform aims to streamline administrative processes, automate routine tasks, and facilitate absolute communication between patients, medical staff, and administrative personnel. The proposed website leverages advanced generative AI approaches to tackle crucial administrative challenges. Through natural language processing (NLP) and image analysis, the system interprets unstructured data such as medical notes, images, and documents. It enables automated documentation generation, medical coding and billing, appointment scheduling, and patient inquiries through a responsive and user-friendly interface

An Interview System Using AI Technology

The rapid advancement of artificial intelligence (AI) has led to innovative applications across various industries, and one such application is the development of an AI-based interview system. This system aims to enhance and streamline the traditional interview process by leveraging AI technologies to assess and interact with candidates. The AI- based interview system incorporates natural language processing (NLP), machine learning (ML), and sentiment analysis algorithms to analyze both verbal and non-verbal cues during interviews. By employing AI, the system can assess a candidate's communication skills, problem-solving abilities, and cultural fit within the organization. The system's machine learning algorithms continuously learn and improve over time, enhancing their ability to accurately evaluate candidates and predict their potential success within the company. The benefits of an AI-based interview system include increased efficiency in the hiring process, reduction in human bias, and the ability to handle a large number of interviews simultaneously. Furthermore, the system can provide data-driven insights to recruiters, aiding in evidence-based decision-making. While the development and implementation of an AI-based interview system pose ethical considerations and challenges related to privacy and bias, ongoing research and ethical guidelines are essential to address these concerns and ensure fair and transparent use of the technology. As technology continues to evolve, further research and collaboration between AI developers and human resource professionals are crucial to harness the full potential of this innovative tool in the hiring process.

Factors associated with social determinants of health mentions in PubMed clinical case reports from 1975 to 2022: A natural language processing analysis

Social determinants of health (SDoH) significantly influence health outcomes, accounting for nearly 40% of such outcomes globally. These determinants, pivotal in understanding health disparities, are insufficiently documented in clinical settings and academic clinical narratives. To address this gap, we examined clinical case reports from PubMed (1975&amp;ndash;2022) to identify mentions of six specific SDoH, employing a pre-trained named-entity recognition (NER) model from Spark natural language processing (NLP). Multivariate logistic regression was utilized to investigate associations between article characteristics and the documentation of SDoH. From 463,546 reports, 4.4% mentioned SDoH, with race/ethnicity being the most dominant mention. Race/ethnicity was often cited by sub-Saharan African authors (adjusted odds ratio [AOR]: 4.47) and in general medicine (AOR: 2.18). Marital status mentions appeared predominantly in psychiatry (AOR: 2.60) and gynecology (AOR: 2.47). Sexual orientation mentions were correlated with infectious diseases (AOR: 25.00) and varied by authorship regions, with stronger associations observed in South America (AOR: 4.04) and North America (AOR: 2.15), and comparatively weaker associations noted in the Indian subcontinent and the Middle East (AOR: 0.16). Immigrant status mentions were closely related to infectious diseases (AOR: 4.51), gynecology (AOR: 4.25), and certain geographies. Homelessness mentions were more prominent in forensic medicine (AOR: 14.92) and in both infections (AOR: 6.36) and mental disorders (AOR: 5.80). Spiritual belief mentions were more prominent with sub-Saharan authors (AOR: 9.17) and psychiatry (AOR: 7.61). SDoH mentions in medical literature were also determined by the diagnosis, cultural background, and journal type. The limited SDoH registration emphasized their overlooked significance. Disproportionate emphasis on specific relationships, such as sexual orientation with infectious diseases, can perpetuate biases and stereotypes. Innovative tools such as Spark NLP offer promise in advancing research using electronic health records (EHRs), but a standardized approach to SDoH reporting and vigilant AI training is crucial for unbiased health-care analysis.

Linguistically Differentiating Acts and Recalls of Racial Microaggressions on Social Media

In this work, we examine the linguistic signature of online racial microaggressions (acts) and how it differs from that of personal narratives recalling experiences of such aggressions (recalls) by Black social media users. We manually curate and annotate a corpus of acts and recalls fromin-the-wild social media discussions, and verify labels with Black workshop participants. We leverage Natural Language Processing (NLP) and qualitative analysis on this data to classify (RQ1), interpret (RQ2), and characterize (RQ3) the language underlying acts and recalls of racial microaggressions in the context of racism in the U.S. Our findings show that neural language models (LMs) can classify acts and recalls with high accuracy (RQ1) with contextual words revealing themes that associate Blacks with objects that reify negative stereotypes (RQ2). Furthermore, overlapping linguistic signatures between acts and recalls serve functionally different purposes (RQ3), providing broader implications to the current challenges in content moderation systems on social media.

The plasma proteome differentiates the multisystem inflammatory syndrome in children (MIS-C) from children with SARS-CoV-2 negative sepsis

BackgroundThe Multi-System Inflammatory Syndrome in Children (MIS-C) can develop several weeks after SARS-CoV-2 infection and requires a distinct treatment protocol. Distinguishing MIS-C from SARS-CoV-2 negative sepsis (SCNS) patients is important to quickly institute the correct therapies. We performed targeted proteomics and machine learning analysis to identify novel plasma proteins of MIS-C for early disease recognition.MethodsA case-control study comparing the expression of 2,870 unique blood proteins in MIS-C versus SCNS patients, measured using proximity extension assays. The 2,870 proteins were reduced in number with either feature selection alone or with a prior COMBAT-Seq batch effect adjustment. The leading proteins were correlated with demographic and clinical variables. Organ system and cell type expression patterns were analyzed with Natural Language Processing (NLP).ResultsThe cohorts were well-balanced for age and sex. Of the 2,870 unique blood proteins, 58 proteins were identified with feature selection (FDR-adjusted P < 0.005, P < 0.0001; accuracy = 0.96, AUC = 1.00, F1 = 0.95), and 15 proteins were identified with a COMBAT-Seq batch effect adjusted feature selection (FDR-adjusted P < 0.05, P < 0.0001; accuracy = 0.92, AUC = 1.00, F1 = 0.89). All of the latter 15 proteins were present in the former 58-protein model. Several proteins were correlated with illness severity scores, length of stay, and interventions (LTA4H, PTN, PPBP, and EGF; P < 0.001). NLP analysis highlighted the multi-system nature of MIS-C, with the 58-protein set expressed in all organ systems; the highest levels of expression were found in the digestive system. The cell types most involved included leukocytes not yet determined, lymphocytes, macrophages, and platelets.ConclusionsThe plasma proteome of MIS-C patients was distinct from that of SCNS. The key proteins demonstrated expression in all organ systems and most cell types. The unique proteomic signature identified in MIS-C patients could aid future diagnostic and therapeutic advancements, as well as predict hospital length of stays, interventions, and mortality risks.

English Vocabulary Learning Aid System Using Digital Twin Wasserstein Generative Adversarial Network Optimized With Jelly Fish Optimization Algorithm

ABSTRACT Natural Language Processing (NLP) is a technology that permits computers to recognize human languages. Words are the fundamental unit of analysis in deep-level grammatical and semantic analysis. The main goal of NLP is typically word segmentation. Since the machine learning techniques cannot be directly applied to the practical issue of significant structural disparities between various data modalities in a multi-modal context. In this paper, English Vocabulary Learning Aid System Using Digital Twin Wasserstein Generative Adversarial Network Optimized using Jelly Fish Optimization Algorithm is proposed. The problematic of multiple modal data heterogeneity is handled by the feature extraction of Parameterized Local Maximum Synchro squeezing Transform and extract the features such as Phonetic features, sentence length, word embedding’s, part of speech tags, word frequencies, N-grams. Then, the Digital twin Wasserstein generative adversarial network classifies the English vocabulary to easy words, intermediate words, and difficult words. The performance of the proposed EVLS-DtwinWGAN-NLP approach attains 3.101%, 7.12%, 7.73% higher accuracy, 24.13%, 13.04%, 29.51% lower computation Time and 2.292%, 5.365%, 1.551% higher AUC compared with existing methods like Feature extraction and analysis of natural language processing for deep learning English language (EVLS-BiLSTM-NLP), State of art for semantic analysis of natural language processing (EVLS-SA-NLP) respectively.

Early dementia detection with speech analysis and machine learning techniques

This in-depth study journey explores the context of natural language processing and text analysis in dementia detection, revealing their importance in a variety of fields. Beginning with an examination of the widespread and influence of text data. The dataset utilised in this study is from TalkBank's DementiaBank, which is basically a vast database of multimedia interactions built with the goal of examining communication patterns in the context of dementia. The various communication styles dementia patients exhibit when communicating with others are seen from a unique perspective by this specific dataset. Thorough data preprocessing procedures, including cleansing, tokenization, and structuring, are undertaken, with a focus on improving prediction capabilities through the combination of textual and non-textual information in the field of feature engineering. In the subsequent phase, the precision, recall, and F1-score metrics of Support Vector Machines (SVM), K-Nearest Neighbours (KNN), Random Forest, and Artificial Neural Networks (ANN) are assessed. Empirical facts are synthesized using text analysis methods and models to formulate a coherent conclusion. The significance of text data analysis, the revolutionary potential of natural language processing, and the direction for future research are highlighted in this synthesis. Throughout this paper, readers are encouraged to leverage text data to embark on their own adventures in the evolving, data-centric world of dementia detection.

Analysis of natural language processing in engineering documentations based on Schweizer-Sklar prioritized aggregation operators for bipolar complex hesitant fuzzy uncertainty

In the context of fuzzy relations, symmetry refers to a property where the relationship between two elements remains the same regardless of the order in which they are considered. Natural language processing (NLP) in engineering documentation discusses the application of computational methods or techniques to robotically investigate, analyze, and produce natural language information for manufacturing contents. The NLP plays an essential role in dealing with large amounts of textual data normally recovered in engineering documents. In this paper, we expose the idea of a bipolar complex hesitant fuzzy (BCHF) set by combining the bipolar fuzzy set (BFS) and the complex hesitant fuzzy set (CHFS). Further, we evaluate some algebraic and Schweizer-Sklar operational laws under the presence of BCHF numbers (BCHFNs). Additionally, using the above information as well as the idea of prioritized (PR) operators, we derive the idea of BCHF Schweizer-Sklar PR weighted averaging (BCHFSSPRWA) operator, BCHF Schweizer-Sklar PR ordered weighted averaging (BCHFSSPROWA) operator, BCHF Schweizer-Sklar PR weighted geometric (BCHFSSPRWG) operator, and BCHF Schweizer-Sklar PR ordered weighted geometric (BCHFSSPROWG) operator. Basic properties for the above operators are also discussed in detail, such as idempotency, monotonicity, and boundedness. Moreover, we evaluate the best way in which NLP can be applied to engineering documentations with the help of the proposed operators. Therefore, we illustrate the major technique of multi-attribute decision-making (MADM) problems based on these derived operators. Finally, we use some existing operators and try to compare their ranking results with our proposed ranking results to show the supremacy and validity of the investigated theory.

Comparison of novel character relationship network mining and drama character relationship shaping algorithms

This study embarks on an interdisciplinary journey to analyze and compare Character Relationships (CR) in two diverse storytelling mediums – the classic novel “Emma” by Jane Austen and the popular TV series “Friends”. Leveraging a blend of Natural Language Processing (NLP) and advanced video analysis tools, this research “sheds light on” the intricate network of CR within these narratives. This study scientifically analyzes these relationships' complete method, creation, and impact using sentiment analysis, object identification, and story coherence algorithms. Qualitative and quantitative metrics such as precision, recall, F-score, and accuracy assist in explaining character updates. The content and TV exhibit distinct storytelling modes, and this study demonstrates that algorithmic analysis of stories is practical. The findings of this study request to contribute to online studies by focusing on understanding character networks and their vital role in TV storytelling.

How do learners’ content network characteristics evolve in cMOOC from the perspective of clustering and comparing learners?

ABSTRACT The content network of cMOOC learners reflects the interactions between learners and valuable external information, and the learners in connectivist learning are diverse. Therefore, it's necessary to make a clear understanding of learners' participation types and content network characteristics. This study proposed a content network characteristics analysis model that included five indexes and corresponding computing methods. 251 learners were included and their interaction data were analyzed through cluster analysis, natural language processing and complex network analysis. The results demonstrated that there were four types of learners, namely “Connecters”, “Followers”, “Likers”, and “Commentators”. There were differences in the content network characteristics among the four types of learners, and the evolution patterns varied. Overall, the values of content network characteristics of Connectors were greatest, and the values of Likers were the lowest. Except for Likers, the content network characteristics of the other three types of learners all exhibited periodic evolution patterns accompanied by the key teaching events, but they varied in evolution trends, specifically manifesting as a spiral upwards trend for Connecters, a gradual trend for Followers, and a steady trend for Commentators. Finally, this study provides valuable suggestions for cMOOC designers and facilitators to support different types of cMOOC learners.

Extracting sensory experiences and cultural ecosystem services from actively crowdsourced descriptions of everyday lived landscapes

ABSTRACT What cultural ecosystem services (CES) do people perceive in their immediate surroundings, and what sensory experiences are linked to these ecosystem services? And how are these CES and experiences expressed in natural language? In this study, we used data generated through a gamified application called Window Expeditions, where people uploaded short descriptions of landscapes they were able to experience through their windows during the COVID-19 pandemic. We used a combination of annotation, close reading and distant reading using natural language processing and graph analysis to extract CES and sensory experiences and link these to biophysical landscape elements. In total, 272 users contributed 373 descriptions in English across more than 40 countries. Of the cultural ecosystem services, recreation was the most prominently described, followed by heritage, identity and tranquility. Descriptions of sensory experiences focused on the visual but also included auditory experiences and touch and feel. Sensory experiences and cultural ecosystem services varied according to biophysical landscape elements, with, for example, animals being more associated with sound and touch/feel and heritage being more associated with moving objects and the built environment. Sentiments also varied across the senses, with the visual being more strongly associated with positive experiences than other senses. This study showed how a hybrid approach combining manual analysis and natural language processing can be productively applied to landscape descriptions generated by members of the public, and how CES on everyday lived landscapes can be extracted from such data sources.

AI Resume Analyzer using Natural Language Processing

The Intelligent Resume Analyzer is an AI-powered system that analyses candidate resumes and makes recommendations to enhance them. The system takes a candidate's resume as input and processes it using machine learning algorithms to discover areas for improvement. The system's main goal is to assist job searchers in creating resumes that can pass the screening process of job interviews. To extract useful insights from the candidate's resume, the system employs a variety of approaches such as natural language processing, text mining, and sentiment analysis. It examines the candidate's experience, skills, education, and achievements to provide feedback on how to improve the resume. The algorithm also evaluates the resume's layout, language, and spelling to verify it follows industry standards. Both job searchers and recruiters can benefit from the Intelligent Resume Analyzer. Recruiters can use the system to swiftly find the top candidates for a certain job, while job seekers can use it to build resumes that are more likely to pass the screening process of job interviews. Job seekers can improve their chances of getting hired for the positions they want by utilizing the Intelligent Resume Analyzer.

Agendas on Nursing in South Korea Media: Natural Language Processing and Network Analysis of News From 2005 to 2022.

In recent years, Korean society has increasingly recognized the importance of nurses in the context of population aging and infectious disease control. However, nurses still face difficulties with regard to policy activities that are aimed at improving the nursing workforce structure and working environment. Media coverage plays an important role in public awareness of a particular issue and can be an important strategy in policy activities. This study analyzed data from 18 years of news coverage on nursing-related issues. The focus of this study was to examine the drivers of the social, local, economic, and political agendas that were emphasized in the media by the analysis of main sources and their quotes. This analysis revealed which nursing media agendas were emphasized (eg, social aspects), neglected (eg, policy aspects), and negotiated. Descriptive analysis, natural language processing, and semantic network analysis were applied to analyze data collected from 2005 to 2022. BigKinds were used for the collection of data, automatic multi-categorization of news, named entity recognition of news sources, and extraction and topic modeling of quotes. The main news sources were identified by conducting a 1-mode network analysis with SNAnalyzer. The main agendas of nursing-related news coverage were examined through the qualitative analysis of major sources' quotes by section. The common and individual interests of the top-ranked sources were analyzed through a 2-mode network analysis using UCINET. In total, 128,339 articles from 54 media outlets on nursing-related issues were analyzed. Descriptive analysis showed that nursing-related news was mainly covered in social (99,868/128,339, 77.82%) and local (48,056/128,339, 48.56%) sections, whereas it was rarely covered in economic (9439/128,339, 7.35%) and political (7301/128,339, 5.69%) sections. Furthermore, 445 sources that had made the top 20 list at least once by year and section were analyzed. Other than "nurse," the main sources for each section were "labor union," "local resident," "government," and "Moon Jae-in." "Nursing Bill" emerged as a common interest among nurses and doctors, although the topic did not garner considerable attention from the Ministry of Health and Welfare. Analyzing quotes showed that nurses were portrayed as heroes, laborers, survivors of abuse, and perpetrators. The economic section focused on employment of youth and women in nursing. In the political section, conflicts between nurses and doctors, which may have caused policy confusion, were highlighted. Policy formulation processes were not adequately reported. Media coverage of the enactment of nursing laws tended to relate to confrontations between political parties. The media plays a crucial role in highlighting various aspects of nursing practice. However, policy formulation processes to solve nursing issues were not adequately reported in South Korea. This study suggests that nurses should secure policy compliance by persuading the public to understand their professional perspectives.

Advanced Natural Language Processing Analysis on Cross-Border Media Sentiment from China and South Korea

Advanced Natural Language Processing Analysis on Cross-Border Media Sentiment from China and South Korea

Text‐based sentiment analysis in finance: Synthesising the existing literature and exploring future directions

SummaryAdvances in Deep Learning have drastically improved the abilities of Natural Language Processing (NLP) research, creating new state‐of‐the‐art benchmarks. Two research streams at the forefront of NLP analysis are transformer architecture and multimodal analysis. This paper critically evaluates the extant literature applying sentiment analysis techniques to the financial domain. We classify the financial sentiment analysis literature according to the most used techniques in the area, with a focus on methods used to detect sentiment within corporate earnings conference calls, because of their dual modality (text‐audio) nature. We find that the financial literature follows a similar path to NLP sentiment literature, in that more advanced techniques to define sentiment are being used as the field progresses. However, techniques used to determine financial sentiment currently fall behind state‐of‐the‐art techniques used within NLP. Two future directions stem from this paper. Firstly, we propose that the adoption of transformer architecture to create robust representations of textual data could enhance sentiment analysis in academic finance. Secondly, the adoption of multimodal classifiers in finance represents a new, currently underexplored area of study that offers opportunities for finance research.

Quantum Vision Transformers

In this work, quantum transformers are designed and analysed in detail by extending the state-of-the-art classical transformer neural network architectures known to be very performant in natural language processing and image analysis. Building upon the previous work, which uses parametrised quantum circuits for data loading and orthogonal neural layers, we introduce three types of quantum transformers for training and inference, including a quantum transformer based on compound matrices, which guarantees a theoretical advantage of the quantum attention mechanism compared to their classical counterpart both in terms of asymptotic run time and the number of model parameters. These quantum architectures can be built using shallow quantum circuits and produce qualitatively different classification models. The three proposed quantum attention layers vary on the spectrum between closely following the classical transformers and exhibiting more quantum characteristics. As building blocks of the quantum transformer, we propose a novel method for loading a matrix as quantum states as well as two new trainable quantum orthogonal layers adaptable to different levels of connectivity and quality of quantum computers. We performed extensive simulations of the quantum transformers on standard medical image datasets that showed competitively, and at times better performance compared to the classical benchmarks, including the best-in-class classical vision transformers. The quantum transformers we trained on these small-scale datasets require fewer parameters compared to standard classical benchmarks. Finally, we implemented our quantum transformers on superconducting quantum computers and obtained encouraging results for up to six qubit experiments.

Sentiment Analysis of Finnish Twitter Discussions on COVID-19 During the Pandemic

With the outbreak of the COVID-19 pandemic, researchers have studied how people reacted on social media during the pandemic. Sentiment analysis has been leveraged to gain insight. However, much of the research conducted on both sentiment analysis and social media analysis of COVID-19 often focuses on widespread languages, such as English and Chinese. This is partly due to the scarcity of resources for natural language processing and sentiment analysis for morphologically complex and less prevalent languages such as Finnish. This paper aims to analyze sentiments on Twitter in the Finnish language during the COVID-19 pandemic. We manually annotate with sentiments a random sample of 1943 tweets about COVID-19 in Finnish. We use it to build binomial and multinomial logistic regression models with Lasso penalty by exploiting ngrams and two existing sentiment lexicons. We also build two similar models using an existing (pre-COVID-19) Twitter dataset for comparison. The best-performing model for the Finnish language is then used to determine the trends of positive, negative, and neutral opinions on a collection of tweets in Finnish extracted between April 21 and June 18, 2020. The best sentiment polarity prediction model for the Finnish language attain 0.785 AUC, 0.710 balanced accuracy, and 0.723 macro-averaged F1 for predicting positive and negative polarity (binomial classification), and 0.667 AUC, 0.607 balanced accuracy, and 0.475 F1 when adding neutral tweets (multinomial classification). On the other hand, the pre-COVID-19 model trained on the same number of tweets exhibits higher accuracy for the multinomial model (0.687 balanced accuracy, and 0.588 F1). We hypothesize that this loss of performance is due to the COVID-19 context that makes sentiment analysis of neutral tweets more difficult for the machine learning algorithm to predict. Running the model on all the extracted Finnish tweets, we observe a decrease in negativity and an increase in positivity over the observed time as the Finnish government lifts restrictions. Our results show that applying an existing general-purpose sentiment analyzer on tweets that are domain-specific, such as COVID-19, provides lower accuracy. More effort in the future needs to be invested in using and developing sentiment analysis tools tailored to their application domain when conducting large-scale social media analysis of specific medical issues, such as a global pandemic.