Word Count Research Articles

Variation in Countability Properties and Noun Classes

This paper establishes that Sorani Kurdish has a mass/count distinction that can be distinguished by looking at, which nouns occur with plural moprhology (-an), optional classifiers (e.g. dane, CLinainmate), unit denumerators (yek, ‘one’), fuzzy denumerators (nsikay panja ‘about fifty’), and other denumerators (herdû ‘both’). Moreover, properties of countability in Sorani Kurdish can be used to distinguish six classes of nouns that differ with respect to their acceptability with the aforementioned properties, examples of nouns in each class being setyare (‘car’), polîs (‘police’), miqes (‘scissors’), manga (‘cattle’), çiyakanî Himalaya (‘Himalayas’), and xîn (‘blood’). This data was gathered by comparing Sorani Kurdish near equivalents to English countability properties found in previous studies, and asking five speakers of Sorani Kurdish for their judgments, following semantic fieldwork methodology. This research expands what is found about countability in Sorani Kurdish grammar books. This paper shows that Sorani Kurdish has both a rich classifier system and mass/count distinction, a combination that has not yet been documented. Knowing these characteristics of Sorani Kurdish helps to create a clearer picture of the linguistic variation regarding countability and suggests a relationship between the number of countability properties and countability classes in a given language.

A Study on the Correlation Between LIWC Word Categories and Chinese Composition Writing Performance of Fourth, Fifth, and Sixth Grade Students

This study focuses on the analysis of the Chinese composition writing performance of fourth, fifth, and sixth grade students in 16 selected schools in Longhua District, Shenzhen during the spring semester of 2023. Using LIWC (Linguistic Inquiry and Word Count) as a text analysis tool, the study explores the impact of LIWC categories on writing performance which is scaled by score. The results show that the simple LIWC word categories have a significant positive influence on the composition scores of lower-grade students; while complex LIWC word categories have a significant negative influence on the composition scores of lower-grade students but a significant positive influence on the composition scores of higher-grade students. Process word categories have a positive influence on the composition scores of all three grades, but the impact of complex process word categories increases as the grade level rises.

Doctor Versus Artificial Intelligence: Patient and Physician Evaluation of Large Language Model Responses to Rheumatology Patient Questions in a Cross-Sectional Study.

The objective of the current study was to assess the quality of large language model (LLM) chatbot versus physician-generated responses to patient-generated rheumatology questions. We conducted a single-center cross-sectional survey of rheumatology patients (n=17) in Edmonton, Alberta, Canada. Patients evaluated LLM chatbot versus physician-generated responses for comprehensiveness and readability, with four rheumatologists also evaluating accuracy by using a Likert scale from 1 to 10 (1 being poor, 10 being excellent). Patients rated no significant difference between artificial intelligence (AI) and physician-generated responses in comprehensiveness (mean 7.12 ± SD 0.99 vs 7.52 ± 1.16; P = 0.1962) or readability (7.90 ± 0.90 vs 7.80 ± 0.75; P = 0.5905). Rheumatologists rated AI responses significantly poorer than physician responses on comprehensiveness (AI 5.52 ± 2.13 vs physician 8.76 ± 1.07; P < 0.0001), readability (AI 7.85 ± 0.92 vs physician 8.75 ± 0.57; P = 0.0003), and accuracy (AI 6.48 ± 2.07 vs physician 9.08 ± 0.64; P < 0.0001). The proportion of preference to AI- versus physician-generated responses by patients and physicians was 0.45 ± 0.18 and 0.15 ± 0.08, respectively (P = 0.0106). After learning that one answer for each question was AI generated, patients were able to correctly identify AI-generated answers at a lower proportion compared to physicians (0.49 ± 0.26 vs 0.97 ± 0.04; P = 0.0183). The average word count of AI answers was 69.10 ± 25.35 words, as compared to 98.83 ± 34.58 words for physician-generated responses (P = 0.0008). Rheumatology patients rated AI-generated responses to patient questions similarly to physician-generated responses in terms of comprehensiveness, readability, and overall preference. However, rheumatologists rated AI responses significantly poorer than physician-generated responses, suggesting that LLM chatbot responses are inferior to physician responses, a difference that patients may not be aware of.

Unfolding the Transitions in Sustainability Reporting

The sustainable development goals (SDGs) have been widely embraced by organizations as a sign of their commitment to sustainability. In this study, we develop a novel SDG-related bidirectional encoder representations from transformers (BERT) model, using the neural network methodology, to determine the thematic evolution of European banks’ sustainability reports. We train this model on the OSDG-CD corpus, which we extend by labeling approximately 10,000 sentences based on SDGs content. The classification capabilities of this model appear to be very effective. Analysts who use our methodology can make faster decisions about the sustainability claims of financial institutions. Our methodology can be extended to non-financial entities. By analyzing the sustainability reports of 98 listed banks covering the accounting periods ranging from 2010 to 2022, we can identify the temporal emphasis of the SDGs. By 2022, climate action had emerged as the most important focus theme. We further validate our classification methodology by establishing a strong correlation between the evolution of SDG prevalence and relevant macroeconomic indicators. We also reveal a difference in focus between various European regions. Finally, we use word counts and k-means cluster analysis to document changes in the objectives of banks by investigating their discussion content.

Tracing the Footprints of AI in Radiology Literature: A Detailed Analysis of Journal Abstracts.

To assess and compare the probabilities of AI-generated content within scientific abstracts from selected Q1 journals in the fields of radiology, nuclear medicine, and imaging, published between May and August 2022 and May and August 2023. An extensive list of Q1 journals was acquired from Scopus in the fields of radiology, nuclear medicine, and imaging. All articles in these journals were acquired from the Medline databases, focusing on articles published between May and August in 2022 and 2023. The study specifically compared abstracts for limitations of the AI detection tool in terms of word constraints. Extracted abstracts from the two different periods were categorized into two groups, and each abstract was analyzed using the AI detection tool, a system capable of distinguishing between human and AI-generated content with a validated accuracy of 97.06 %. This tool assessed the probability of each abstract being AI-generated, enabling an in-depth comparison between the two groups in terms of the prevalence of AI-generated content probability. Group 1 and Group 2 exhibit significant variations in the characteristics of AI-generated content probability. Group 1, consisting of 4,727 abstracts, has a median AI-generated content probability of 3.8 % (IQR1.9-9.9 %) and peaks at 49.9 %, with the computation times contained within a range of 2 to 10seconds (IQR 3-8 s). In contrast, Group 2, which is composed of 3,917 abstracts, displays a significantly higher median AI-generated content probability at 5.7 % (IQR2.8-12.9 %) surging to a maximum of 69.9 %, with computation times spanning from 2 to 14 seconds (IQR 4-11 s). This comparison yields a statistically significant difference in median AI-generated content probability between the two groups (p = 0.005). No significant correlation was observed between word count and AI probability, as well as between article type, primarily original articles and reviews, and AI probability, indicating that AI probability is independent of these factors. The comprehensive analysis reveals significant differences and variations in AI-generated content probabilities between 2022 and 2023, indicating a growing presence of AI-generated content. However, it also illustrates that abstract length or article type does not impact the likelihood of content being AI-generated. · The study examines AI-generated content probability in scientific abstracts from Q1 journals between 2022 to 2023.. · The AI detector tool indicates an increased median AI content probability from 3.8 % to 5.7 %.. · No correlation was found between abstract length or article type and AI probability..

Noun classifiers in Hong Kong Sign Language

Abstract As most other sign languages, Hong Kong Sign Language (HKSL) makes use of verbal classifiers for the purpose of spatial predication. However, a thorough study of the Asian SignBank shows that these same classifier handshapes occur in 76% of all HKSL nouns. In this paper, I argue that these classifier handshapes embedded in nouns are in fact noun classifiers. Under the framework of Distributed Morphology, classifier handshapes are underspecified roots in root compounds which can be assigned the nominal category when merged with a nominal functional head in syntax. The function of these classifiers is to identify, and categorize a discourse referent, and they are full-fledged classifiers according to standards set by the general linguistic literature on classifiers. In addition to verbal and noun classifiers, I also briefly show how HKSL uses mensural numeral classifiers to intervene between mass nouns and numerals within the NP. Taken together, this means that HKSL employs three different types of classifiers: verbal, noun, and numeral classifiers.

Linguistic Variables and Gender Differences Within a Messenger-Based Psychosocial Chat Counseling Service for Children and Adolescents: Cross-Sectional Study.

Text messaging is widely used by young people for communicating and seeking mental health support through chat-based helplines. However, written communication lacks nonverbal cues, and language usage is an important source of information about a person's mental health state and is known to be a marker for psychopathology. The aim of the study was to investigate language usage, and its gender differences and associations with the presence of psychiatric symptoms within a chat counseling service for adolescents and young adults. For this study, the anonymized chat content of a German messenger-based psychosocial chat counseling service for children and adolescents ("krisenchat") between May 2020 and July 2021 was analyzed. In total, 661,131 messages from 6962 users were evaluated using Linguistic Inquiry and Word Count, considering the following linguistic variables: first-person singular and plural pronouns, negations, positive and negative emotion words, insight words, and causation words. Descriptive analyses were performed, and gender differences of those variables were evaluated. Finally, a binary logistic regression analysis examined the predictive value of linguistic variables on the presence of psychiatric symptoms. Across all analyzed chats, first-person singular pronouns were used most frequently (965,542/8,328,309, 11.6%), followed by positive emotion words (408,087/8,328,309, 4.9%), insight words (341,460/8,328,309, 4.1%), negations (316,475/8,328,309, 3.8%), negative emotion words (266,505/8,328,309, 3.2%), causation words (241,520/8,328,309, 2.9%), and first-person plural pronouns (499,698/8,328,309, 0.6%). Female users and users identifying as diverse used significantly more first-person singular pronouns and insight words than male users (both P<.001). Negations were significantly more used by female users than male users or users identifying as diverse (P=.007). Similar findings were noted for negative emotion words (P=.01). The regression model of predicting psychiatric symptoms by linguistic variables was significant and indicated that increased use of first-person singular pronouns (odds ratio [OR] 1.05), negations (OR 1.11), and negative emotion words (OR 1.15) was positively associated with the presence of psychiatric symptoms, whereas increased use of first-person plural pronouns (OR 0.39) and causation words (OR 0.90) was negatively associated with the presence of psychiatric symptoms. Suicidality, self-harm, and depression showed the most significant correlations with linguistic variables. This study highlights the importance of examining linguistic features in chat counseling contexts. By integrating psycholinguistic findings into counseling practice, counselors may better understand users' psychological processes and provide more targeted support. For instance, certain linguistic features, such as high use of first-person singular pronouns, negations, or negative emotion words, may indicate the presence of psychiatric symptoms, particularly among female users and users identifying as diverse. Further research is needed to provide an in-depth look into language processes within chat counseling services.

The Association Between Linguistic Characteristics of Physicians' Communication and Their Economic Returns: Mixed Method Study.

Web-based health care has the potential to improve health care access and convenience for patients with limited mobility, but its success depends on active physician participation. The economic returns of internet-based health care initiatives are an important factor that can motivate physicians to continue their participation. Although several studies have examined the communication patterns and influences of web-based health consultations, the correlation between physicians' communication characteristics and their economic returns remains unexplored. This study aims to investigate how the linguistic features of 2 modes of physician-patient communication, instrumental and affective, determine the physician's economic returns, measured by the honorarium their patients agree to pay per consultation. We also examined the moderating effects of communication media (web-based text messages and voice messages) and the compounding effects of different communication features on economic returns. We collected 40,563 web-based consultations from 528 physicians across 4 disease specialties on a large, web-based health care platform in China. Communication features were extracted using linguistic inquiry and word count, and we used multivariable linear regression and K-means clustering to analyze the data. We found that the use of cognitive processing language (ie, words related to insight, causation, tentativeness, and certainty) in instrumental communication and positive emotion-related words in affective communication were positively associated with the economic returns of physicians. However, the extensive use of discrepancy-related words could generate adverse effects. We also found that the use of voice messages for service delivery magnified the effects of cognitive processing language but did not moderate the effects of affective processing language. The highest economic returns were associated with consultations in which the physicians used few expressions related to negative emotion; used more terms associated with positive emotions; and later, used instrumental communication language. Our study provides empirical evidence about the relationship between physicians' communication characteristics and their economic returns. It contributes to a better understanding of patient-physician interactions from a professional-client perspective and has practical implications for physicians and web-based health care platform executives.

Alexithymia in the Narratization of Romantic Relationships: The Mediating Role of Fear of Intimacy.

The purpose of the study was to verify the hypothesis concerning the relationship between alexithymia and selected indicators used to describe emotional events, specifically romantic relationships. Alexithymia, due to significant distortions in cognitive processing of emotional content, is demonstrated by poor recognition of emotions in oneself and others and, as a result, by deficits in empathy, avoidance of social relationships, and deficits in the ability to mentalize. Differences in narrations were tested by alexithymia levels (high vs. low) and the relation between specific narration features and individual alexithymia factors, i.e., difficulties in identifying emotions, difficulties in verbalising emotions, and externally oriented thinking. A total of 356 people who had been in a romantic relationship for at least six months participated in the study. The TAS-20 was applied to measure alexithymia, and the FIS questionnaire was used to investigate anxiety in close relationships. Participants were asked to freely describe the romantic relationship they were in at that moment. The Linguistic Inquiry Word Count (LIWCLIWC2015 v1.6-unlimited duration academic licence) software was used for content analysis. The study was conducted online. On the basis of the analyses conducted, high levels of alexithymia were found to be significantly associated with a lower total number of words used in narrative, a lower number relating to positive emotions, a lower number relating to causation and insight, and a higher number relating to negative emotions. Various results were obtained for individual dimensions of alexithymia in relation to the LIWC categories and the mediating role of fear of intimacy. For the difficulty identifying feelings (DIF), a significant mediating effect was observed only for words associated with negative emotions, whereas for the difficulty describing feelings (DDF), significant mediating effects were found for words relating to negative emotions and causality. In the case of externally oriented thinking (EOT), significant mediating effects were obtained for all analysed categories from LIWC.

In their own words: linguistic analysis of interviews with Sesotho mothers on their child’s first thousand days

ABSTRACT Sensitive and responsive caregiving is a critical protective factor for children in Sub-Saharan Africa (SSA) during their first 1000 days of life. Health initiatives may benefit from evaluating the experience of caregivers in SSA to understand their parenting experiences. This study utilized linguistic analysis to gain insight into the lived experience of 19 Sesotho-speaking mothers of infants aged 18–30 months in low-income areas of Mangaung, Free State, South Africa. Women were asked the same set of questions about their child’s first 1000 days of life, and we used LIWC (Linguistic Inquiry and Word Count) LIWC to evaluate the perspectives of mothers in SA compared to published LIWC norms. Analyses revealed word use patterns associated with psychological processes such as negative affect, social processes, anger and anxiety. Findings are discussed within the context of Sesotho culture to aid in the future intervention development for this vulnerable population. This study is important because it yields important information about the subjective experience of an understudied critical population of mothers in SSA. Further, this study points to future directions for the field of understanding the impact of motherhood on women in low resource settings.

Association between patients’ word use and symptom reduction in an internet-based cognitive behavioral intervention for prolonged grief symptoms: A secondary analysis of a randomized controlled trial

Bereaved by suicide face unique challenges and have differences in their language compared to bereaved by other causes of death, however their language during therapy has not been studied yet. This study investigates the association between patients’ language and reduction in prolonged grief symptoms in an internet-based intervention for people bereaved by suicide. Data stems from a randomized controlled trial including 47 people completing self-reported surveys. Patient language was analyzed using the Linguistic Inquiry and Word Count program. Symptom change was determined through absolute change scores. Stepwise forward regression and repeated measures analyses of variances were calculated. During confrontation, a higher reduction of prolonged grief symptoms was predicted by more words describing perceptual (β = − 0.43, p = .002) and cognitive processes (β = − 0.63, p = .002) and less present focus words (β = 0.66, p = .002). During cognitive restructuring, more words describing drives (β = − 0.40, p = .004), less past focus words (β = 0.59, p = .002) and less informal language (β = 0.40, p = .01) predicted a higher reduction of prolonged grief symptoms. Lastly, during behavioral activation, more past focus words (β = − 0.54, p = .002) predicted a higher grief reduction. Findings underline the importance of exposure and cognitive restructuring during therapy and further suggest the relevance of the previously not studied linguistic perceptual processes. Moreover, this study emphasizes the importance of different tenses throughout the intervention, adding knowledge to previous studies assessing time at a single point in therapy.

Fight against hair loss together: exploring self-disclosure and social support in an online hair loss support community

Purpose Hair loss is often overlooked but psychologically challenging. However, the emergence of online health communities provides opportunities for hair loss patients to seek social support through self-disclosure. Nevertheless, not all disclosures receive the desired support. This research explores what patients disclose within the community and how their health narrative (content, form and linguistic style) regarding self-disclosure influences the social support they receive.Design/methodology/approachThis study investigated a 13-year-old online support group for Chinese hair loss patients with nearly 240,000 members. Using structural topic modeling, Linguistic Inquiry and Word Count, and a negative binomial model, the research analyzed the content of self-disclosure and the interrelationships between social support and three narrative dimensions of self-disclosure.FindingsSelf-disclosures are classified into 14 topics, grouped under analytical, informative and emotional categories. Emotion-related self-disclosures, whether in content or effective word use, receive deeper social support. Longer and image-rich posts attract more support in quantity, but not necessarily in quality, while cognitive words have a limited impact.Originality/valueThis study addresses the previously overlooked population of hair loss patients within online health communities. It employs a more comprehensive health narrative framework to explore the relationship between self-disclosure and social support, utilizing unsupervised structural topic modeling methods to mine text. The research offers practical implications for how patients seek support and for healthcare professionals in developing doctor-patient communication strategies.

The Ground Floor Approach to Video Game Accessibility: Identifying Design Features Prioritized by Accessibility Reviews

This article describes the priorities for accessibility found in video game accessibility reviews. Through an analysis of 20 articles published by the accessibility review websites DAGERSystem.com and CanIPlayThat.com, this study categorizes the accessibility features in those articles in order to discover which accessibility features are the most important to reviewers, gauged by how many words are dedicated to addressing each feature. The study discovers 15 primary categories of accessibility most prominently features in these reviews. Word counts for each category provide a prioritized list of which accessibility features are the most central to video game accessibility. This article concludes by introducing the ground floor approach to video game accessibility, which calls for subtitles, difficulty settings, control options, and visual clarity as essential features in modern game design.

An evaluation of orthodontic information quality regarding artificial intelligence (AI) chatbot technologies: A comparison of ChatGPT and google BARD

Abstract Introduction In recent times, chatbots have played an increasing and noteworthy role in the field of medical practice. The present research was conducted to evaluate the accuracy of the responses provided by ChatGPT and BARD, two of the most utilised chatbot programs, when interrogated regarding orthodontics. Materials and methods Twenty-four popular questions about conventional braces, clear aligners, orthognathic surgery, and orthodontic retainers were chosen for the study. When submitted to the ChatGPT and Google BARD platforms, an experienced orthodontist and an orthodontic resident rated the responses to the questions using a five-point Likert scale, with five indicating evidence-based information, four indicating adequate information, three indicating insufficient information, two indicating incorrect information, and one indicating no response. The results were recorded in Microsoft Excel for comparison and analysis. Results No correlation was found between the ChatGPT and Google BARD scores and word counts. However, a moderate to significant relationship was observed between the scores and several listed references. No significant association was found between the number of words and references, and a statistically significant difference was observed in both investigators’ numerical rating scales using the AI tools (p = 0.014 and p = 0.030, respectively). Conclusion Generally, ChatGPT and BARD provide satisfactory responses to common orthodontic inquiries that patients might ask. ChatGPT’s answers marginally surpassed those of Google BARD in quality.

Evaluating Learners' Online Learning Experience of Informal Learning Environments: A LIWC Analysis

Online video-based content, such as massive open online courses (MOOCs) or educational videos on YouTube, are popular ways for individuals to learn outside of a traditional learning space. Video-based informal learning environments fit in the framework of community of inquiry (CoI), with unique teaching, cognitive, and social presence. Videos on the environment provide direct instruction representing teaching presence, and the feature of online discussion allows learners to comment on instructional materials and interact with other learners, which show social and cognitive presence. Multi-format informal learning environments bring learners diverse learning experiences and analyzing the content of online discussions can help us better understand learners’ experiences. This study uses Linguistic Inquiry and Word Count (LIWC) to evaluate learners' learning experiences on YouTube and edX through the construct of CoI. After matching LIWC’s keywords with social presence and cognitive presence, we analyze learners’ comments (N = 6,938) collected from YouTube and edX. Our findings reveal that there is a more differentiated cognitive presence and a more similar social presence between YouTube and edX.

Review and Evaluation of European National Clinical Practice Guidelines for the Treatment and Management of Active Charcot Neuro-Osteoarthropathy in Diabetes Using the AGREE-II Tool Identifies an Absence of Evidence-Based Recommendations.

Background: Charcot neuro-osteoarthropathy (CNO) is a rare but devastating complication of diabetes associated with high rates of morbidity; yet, many nonfoot specialists are unaware of it, resulting in missed and delayed diagnosis. Clinical practice guidelines (CPGs) have proven useful in improving quality of care and standardizing practice in diabetes and diabetic foot care. However, little is known about the consistency in recommendations for identification and management of active CNO. Aim: The aim of this study is to review European national diabetes CPGs for the diagnosis and management of active CNO and to assess their methodological rigor and transparency. Methods: A systematic search was performed to identify diabetes national CPGs across Europe. Guidelines in any language were reviewed to explore whether they provided a definition for active CNO and recommendations for diagnosis, monitoring, and management. Methodological rigor and transparency were assessed using the Appraisal of Guidelines for Research and Evaluation (AGREE-II) tool, which comprises 23 key items organized within six domains with an overall guideline assessment score of ≥ 60% considered to be of adequate quality to recommend use. Each guideline was assessed by two reviewers, and inter-rater agreement (Kendall's W) was calculated for AGREE-II scores. Results: Seventeen CPGs met the inclusion criteria. Breadth of CNO content varied across guidelines (median (IQR) word count: 327; Q1 = 151; Q3 = 790), and 53% provided a definition for active CNO. Recommendations for diagnosis and monitoring were provided by 82% and 53%, respectively, with offloading being the most common management recommendation (88%). Four guidelines (24%) reached threshold for recommendation for use in clinical practice (≥ 60%) with the scope and purpose domain scoring highest (mean (SD): 67%, ± 23%). The remaining domains had average scores ranging between 19% and 53%. Inter-rater agreement was strong (W = 0.882; p < 0.001). Conclusions: European national CPGs for diabetes provide limited recommendations on active CNO. All guidelines showcased deficits in their methodology, suggesting that more rigorous methods should be employed for diabetes CPG development across Europe.

Հայաստանի ազգային գրադարանի նյութերի թվայնացման նոր փուլը (OCR)

Optical character recognition (OCR) is the process of converting an image of text into a machine-encoded text format. For example, when we scan a form or document, the computer saves the scan as an image file. A text editor cannot be used to edit, search, or count words in an image file. OCR helps convert an image into a text document and content is stored as text data. Optical character recognition (OCR) provides the added functionality of being able to edit and search these documents. Such data can be used for analysis, optimization, process automation and productivity improvement. It saves time and reduces errors. We inform our thousands of users that the OCR-passed digital collections of old and new digitized periodicals, editions dedicated to the anniversaries of periodicals are replenished thanks to daily hard work. As of 26.10.2023 #armocr #ocr has received around 93,691 PDFs or 1,496,939 pages: newspaper, magazine, abstracts, NLA publication. These repositories can serve as a unique online library of educational resources.

Detecting Deception Using MCI in Twitter

The purpose of the study was to test whether Modified Cognitive Interviewing (MCI) is an effective method for detecting deceptive human eyewitness accounts in a computer-mediated computer platform with limited text space (such as in Twitter). 44 college students were the participants in this study, where they either had to perform, or pretend that they had performed a cognitive task. Of the 44 participants, 15 performed the task and reported truthfully about their activities; 16 performed the task and denied having participated in the task; 13 read the instructions about the cognitive task and when interviewed claimed to have actually performed the task. The transcripts of interviews, conducted in Twitter, were rated by individuals trained in cognitive interviewing; forensic speech variables (response length (RL), unique word (UW) count and type-token ratio (TTR) were coded from transcripts. Human rater judgments and computer-based speech analysis performed better than chance; computer based judgments were superior to the human judgments (i.e., 79% vs. 54%, respectively). Speech content variables derived from MCI differed significantly, and in different ways, between the truthful and false claimant participants and also between the truthful and denial type participants. MCI derived statement analysis methods are a scientifically valid method, when used in Twitter, that can be used by professionals tasked with distinguishing between true claims, false claims and denials.

A Data-Driven Framework for Power System Event Type Identification via Safe Semi-Supervised Techniques

This paper investigates the use of phasor measurement unit (PMU) data with deep learning techniques to construct real-time event identification models for transmission networks. Increasing penetration of distributed energy resources represents a great opportunity to achieve decarbonization, as well as challenges in systematic situational awareness. When high-resolution PMU data and sufficient manually recorded event labels are available, the power event identification problem is defined as a statistical classification problem that can be solved by numerous cutting-edge classifiers. However, in real grids, collecting tremendous high-quality event labels is quite expensive. Utilities frequently have a large number of event records without in-depth details (i.e., unlabeled events). To bridge this gap, we propose a novel semi-supervised learning-based method to improve the performance of event classifiers trained with a limited number of labeled events by exploiting the information from massive unlabeled events. In other words, compared to existing data-driven methods, our method requires only a small portion of labeled data to achieve a similar level of accuracy. Meanwhile, this work discusses and addresses the performance degradation caused by class distribution mismatch between the training set and the real applications. Specifically, this method utilizes pseudo-labeling technique to investigate the value of unlabeled events and incrementally expands the training dataset. Moreover, a safe learning mechanism is developed to mitigate the impacts of class distribution mismatch and prevent performance degradation. Based on the proposed safe learning mechanism, our model does not directly use all unlabeled events during model training, but selectively uses them through a comprehensive evaluation procedure. Numerical studies on a sizable PMU dataset have been used to validate the performance of the proposed method.

ChatGPT v4 outperforming v3.5 on cancer treatment recommendations in quality, clinical guideline, and expert opinion concordance.

To assess the quality and alignment of ChatGPT's cancer treatment recommendations (RECs) with National Comprehensive Cancer Network (NCCN) guidelines and expert opinions. Three urologists performed quantitative and qualitative assessments in October 2023 analyzing responses from ChatGPT-4 and ChatGPT-3.5 to 108 prostate, kidney, and bladder cancer prompts using two zero-shot prompt templates. Performance evaluation involved calculating five ratios: expert-approved/expert-disagreed and NCCN-aligned RECs against total ChatGPT RECs plus coverage and adherence rates to NCCN. Experts rated the response's quality on a 1-5 scale considering correctness, comprehensiveness, specificity, and appropriateness. ChatGPT-4 outperformed ChatGPT-3.5 in prostate cancer inquiries, with an average word count of 317.3 versus 124.4 (p < 0.001) and 6.1 versus 3.9 RECs (p < 0.001). Its rater-approved REC ratio (96.1% vs. 89.4%) and alignment with NCCN guidelines (76.8% vs. 49.1%, p = 0.001) were superior and scored significantly better on all quality dimensions. Across 108 prompts covering three cancers, ChatGPT-4 produced an average of 6.0 RECs per case, with an 88.5% approval rate from raters, 86.7% NCCN concordance, and only a 9.5% disagreement rate. It achieved high marks in correctness (4.5), comprehensiveness (4.4), specificity (4.0), and appropriateness (4.4). Subgroup analyses across cancer types, disease statuses, and different prompt templates were reported. ChatGPT-4 demonstrated significant improvement in providing accurate and detailed treatment recommendations for urological cancers in line with clinical guidelines and expert opinion. However, it is vital to recognize that AI tools are not without flaws and should be utilized with caution. ChatGPT could supplement, but not replace, personalized advice from healthcare professionals.