Search Scope Research Articles

The Use of AI In Digital Transformation Ethics and Public Relations Strategies

The use of artificial intelligence (AI) has become a significant trend in various sectors, including in public relations practice. This research aims to examine the ethics and strategies of using artificial intelligence in public relations practice. The research method used in this research is qualitative, which is based on literature review studies; data was collected from the Sinta Journal, Scopus, and Garuda Ristek Dikti databases. The scope and focus of the article search apply explicitly to the topic of ethics and Artificial Intelligence-based strategies using search queries based on the Google Scholar search engine. Supporting data comes from books and other relevant reference sources. The research results highlight the importance of maintaining balance and diversity of data and maintaining its confidentiality by applicable privacy regulations. Strategically, AI can help parse text to accurately identify public sentiment, improve interactions withchatbots and virtual assistants, and use recommendation systems based on browsing history or user interactions. The practical implications of this research emphasize the need for PR practitioners to develop transparent and responsible policies for managing data processed by AI, as well as ensuring its use is in line with organizational communication goals and to improve relations with the public. There are also opportunities for further research into the ethical and strategic implications of the use of AI in PR, as well as the development of more sophisticated methods for applying this technology in organizational communications contexts.

Different minimally invasive surgical methods to hysterectomy for benign gynecological disease: A systematic review and network meta-analysis.

This network meta-analysis aimed to compare the perioperative efficacy of various minimally invasive hysterectomy procedures for treating benign gynecological diseases and to assess whether vaginal natural orifice transluminal endoscopic hysterectomy (VNOTEH), a recently emerging procedure, is inferior to traditional laparoscopy. We searched PubMed, Cochrane Library, Embase, China National Knowledge Infrastructure (CNKI), China Biology Medicine disc (CBM), Wanfang Data, and China VIP Database from inception to August 2022 and updated in June 2023. We included randomized controlled trials (RCTs) comparing different minimally invasive hysterectomy techniques in patients with benign gynecological conditions. The intervention measures included nine minimally invasive hysterectomies. The two researchers used the Cochrane risk-of-bias assessment tool for study appraisal. All statistical analyses and drawings were performed using STATA 17.0 and R 4.4.1. A network meta-analysis (NMA) was conducted to compare the effectiveness of minimally invasive hysterectomy and rank its relative impact probabilistically. A total of 78 RCTs involving 7640 patients and nine minimally invasive hysterectomy methods with 16 intervention combinations were included in this study. Among these, 2, 63, and 13 studies were deemed to have a low, medium, and high risk of bias, respectively. Based on the Surface Under the Cumulative Ranking (SUCRA) probability ranking results of NMA, laparoendoscopic single-site surgery-laparoscopic-assisted vaginal hysterectomy (LESS-LAVH) demonstrated superior outcomes in terms of complications, infections, and 24-h postoperative pain scores. LAVH exhibited better performance in injuries and hospital stays, total laparoscopic hysterectomy showed the least blood loss, and vaginal hysterectomy had the shortest operation time. LESS-LAVH and LAVH are recommended options, if feasible. Meanwhile, VNOTEH can achieve comparable results to traditional laparoscopy but requires careful attention to the risk of injury and infection. Future research should aim to broaden the search scope by including high-quality, large-scale, multicenter RCTs.

A meta-analysis of the relationship between overtime work and neck musculoskeletal disorders in Chinese occupational population

Objective: To systematically evaluate the effect of overtime work on neck musculoskeletal disorders of occupational population in China. Methods: In August 2023, the literatures on the relationship between overtime work and neck musculoskeletal disorders were searched in China National Knowledge Infrasture (CNKI), WanFang Database, Chinese Science & Technology Journal Database (VIP), China Biomedical Literature Service System (SinoMed) and PubMed. The search scope was Chinese and English literatures published before June 30, 2023. The Quality of literature studies was evaluated using the Agency for Healthcare Research and Quality (AHRQ) cross-sectional study quality evaluation criteria. The Stata 15.0 software was used to calculate combined OR values, and subgroup analysis and meta regression were used to explore the sources of heterogeneity. Results: A total of 32 literatures were included, all of which were in Chinese, with a total sample size of 36003 people, and the literature quality scores were 5-7 points. Heterogeneity test showed that there was no heterogeneity between the literatures (I(2)=20.3%, P=0.156), so fixed effect model was selected to calculate the combined OR value. The combined OR value of overtime work on neck musculoskeletal disorders was 1.51 (95%CI: 1.42-1.60, P<0.001). Subgroup analysis showed that the combined OR of occupational population <35 years old group was higher than that in the ≥35 years old group (P<0.05). But there were no significant differences in the combined OR values in the subgroups classified by average years of service, male proportion, type of OR values, sample size and industry classification (P>0.05) . Conclusion: Overtime work is a risk factor for neck musculoskeletal disorders in Chinese occupational population. Working hours should be rationally arranged to reduce the prevalence of neck musculoskeletal disorders.

Evolutionary stochastic configuration networks for industrial data analytics

Stochastic configuration network (SCN) with compact architecture is expected for data modeling. However, the hidden-node parameters (HNPs) randomly configured may result in a slow learning process due to the redundant nodes embedded in the model. To resolve this problem, an evolutionary SCN based on an improved differential evolution (DE) algorithm is proposed in this paper. Specifically, the improved DE reuses the assignment information of last hidden node to find an appropriate search scope for the current one; employs a space reduction method to seed a promising population in the scope; and develops a performance-aware scheme to adjust the scale factor of mutation operators. The proposed evolutionary SCNs are compared with other methods on six datasets and then applied for two real-world applications. Experimental results demonstrate that the proposed method obtains superior performance in terms of compactness and accuracy, with great potential for real-world data analysis.

How high-speed railway expands beyond local knowledge search in interdisciplinary innovation: Evidence from China

Breaking through local knowledge search limitations is essential for fostering interdisciplinary innovation. This paper investigates how High-speed Railway (HSR) facilitates these firms in integrating information technology (IT) knowledge from other cities into their innovation processes. By connecting cities through faster transportation, HSR diminishes the constraints of geographical proximity, enabling firms to bypass the traditional reliance on local knowledge clusters and access a broader pool of IT expertise. Utilizing a difference-in-differences (DID) approach, our results reveal that HSR enhances knowledge search scope by improving access to IT resources, increasing human capital mobility, and raising competitive awareness. Additionally, we identify heterogeneity in this effect based on factors such as the marketization level of a firm's location, the degree of digital infrastructure development, whether a firm is located in a provincial capital, and the specific industry sector. Our findings further demonstrate that the expansion of knowledge search scope due to HSR significantly boosts future firm performance. These findings highlight the critical role of transportation infrastructure, exemplified by HSR, in breaking down geographic barriers to interdisciplinary knowledge flows and fostering more dynamic innovation ecosystems.

A Comparative Bibliometric Analysis of Ethical Leadership Research: Insights From Scopus and Web of Science (2010-2024)

This research provides a bibliometric analysis of articles published between 2010 and 2024 on "Ethical Leadership" using Scopus and Web of Science databases. The analysis aims to see the status of the ethical leadership. Ethical leadership represents a crucial aspect of effective management, and it has been the subject of increasing interest from researchers. The investigation involved an examination of both Scopus and Web of Science. These databases are widely recognized as prominent sources of academic research. The comparative analysis is designed to provide detailed insight into the citation patterns, and collaborative networks associated with ethical leadership research. The search scope was narrowed by using the term "ethical leadership" as a keyword in both databases. Furthermore, other disciplines were excluded by entering "education" into the search filter. The term "Ethical Leadership" was entered into the search field. As a result, 122 articles obtained from the Web of Science database and 133 articles from Scopus. Different results were obtained from the data analyzed.

The Form of the Multinational Enterprises: The Evolution in the Context of the Functional Analysis of Law

In this article, we identify two recent stages in the evolution of the legal form of multinational enterprises (MNEs): the rise of offshore empires by the late 20th century and the emergence of neo-MNEs in the past decade. We examine the gradual shift in the parameters linked to the economic substance unity of MNEs (Part 1) and those associated with their legal multiplicity (Part 2), within the context of the law’s functional analysis. Our findings have revealed that each MNE’s form developmental stage corresponds to changes in all the selected parameters. Furthermore, we note significant changes in the relationship between the MNE’s legal form and its economic substance. This has led us to conclude that the firm’s size is not a necessary characteristic of the MNEs’ form, but serves as a crucial additional criterion for defining the scope of regulatory measures. Simultaneously, optimization of regulatory burdens including the growing importance of sanction compliance is a key factor of forming the MNEs. The differences between new and old MNEs’ forms are primarily the result of technological transformations, reflecting not only a mismatch but also a fluidity in the economic and legal boundaries of MNEs (Part 3). In conclusion, a comprehensive, cross-sectoral approach is required for the legal form of MNEs. Understanding the specifics of neo-MNEs broadens the search scope for new conceptual solutions. States initiate experimental legal regimes, while entrepreneurs develop quasi-public and quasi-corporate structures based on new technologies and quasi-law (standards, customs, soft law). This approach eventually prompts a reevaluation of the normativity concept

Combinatorial prediction of therapeutic perturbations using causally-inspired neural networks.

As an alternative to target-driven drug discovery, phenotype-driven approaches identify compounds that counteract the overall disease effects by analyzing phenotypic signatures. Our study introduces a novel approach to this field, aiming to expand the search space for new therapeutic agents. We introduce PDGrapher, a causally-inspired graph neural network (GNN) designed to predict combinatorial perturbagens - sets of therapeutic targets - capable of reversing disease effects. Unlike methods that learn responses to perturbations, PDGrapher solves the inverse problem, which is to infer the perturbagens necessary to achieve a specific response - i.e., directly predicting perturbagens by learning which perturbations elicit a desired response. By encoding gene regulatory networks or protein-protein interactions, PDGrapher can predict unseen chemical or genetic perturbagens, aiding in the discovery of novel drugs or therapeutic targets. Experiments across nine cell lines with chemical perturbations show that PDGrapher successfully predicted effective perturbagens in up to 13.33% additional test samples and ranked therapeutic targets up to 35% higher than the competing methods, and the method shows competitive performance across ten genetic perturbation datasets. A key innovation of PDGrapher is its direct prediction capability, which contrasts with the indirect, computationally intensive models traditionally used in phenotype-driven drug discovery that only predict changes in phenotypes due to perturbations. The direct approach enables PDGrapher to train up to 25 times faster than methods like scGEN and CellOT, representing a considerable leap in efficiency. Our results suggest that PDGrapher can advance phenotype-driven drug discovery, offering a fast and comprehensive approach to identifying therapeutically useful perturbations.

Hand eczema: a review of clinical, dermoscopic and histological features.

Hand eczema (HE) is a prevalent inflammatory skin condition, impacting 15-20% of individuals, with a notable incidence rate of approximately 7.3 cases per 1000 person/years. This condition exhibits significant gender-based variability, with a higher prevalence in females. The clinical presentation of HE is characterized by pruritic erythematous, edematous, weeping plaques, vesicles, and/or bullae, showcasing considerable heterogeneity. A literature search was conducted across multiple databases, including Medline, Pubmed, Scopus, and the Cochrane Library. The search was conducted using the following key words and medical subject heading (MeSH) terms: "hand," "eczema," "dermatitis," "dermoscopy," and "histology," employing the Boolean term "AND" to combine the research terms for optimal search precision. PRISMA algorithm has been used for article screening. The search scope included manuscripts published up to October 1, 2023. Up to 50% of HE cases are associated with atopic dermatitis, emphasizing the complex interplay between various dermatological conditions. Common subtypes of HE include irritant contact dermatitis (ICD), allergic contact dermatitis (ACD), atopic hand eczema (AHE), and protein contact dermatitis/contact urticaria (PCD). The chronic nature of HE presents a substantial management challenge, often underestimated, leading to delayed treatment and potential progression to chronic hand eczema (CHE). Beyond individual health implications, HE exerts a profound impact on occupational, domestic, social, and psychological aspects, establishing itself as the most prevalent occupation-related skin disease. This paper seeks to establish a comprehensive classification system for HE, integrating clinical, dermoscopic, and histological elements. Dermoscopy, specifically, proves instrumental in distinguishing HE from palmar psoriasis, revealing characteristic features such as yellow scales and irregular vessels. Histopathological findings underscore the dynamic changes observed from acute to chronic stages, while challenges in differentiating hyperkeratotic HE from psoriasis underscore the necessity for a holistic diagnostic approach. Accurate diagnosis and effective management of HE necessitate a holistic perspective that recognizes the inherent complexities of this inflammatory skin disease. By providing a multidimensional classification system, incorporating clinical, dermoscopic, and histological parameters, this paper aimed to contribute to a more nuanced understanding of HE and facilitate improved approaches to its diagnosis and treatment.

An improved whale algorithm for real-time boundary processing and its convergence analysis

Abstract To tackle the deficiencies in the whale algorithm, including its weak stability, sluggish convergence rate, and tendency to get trapped in local optima, an adapted whale optimization algorithm has been created, which introduces quasi-opposition learning and a real-time boundary processing mechanism. In the enhanced algorithm, a quasi-opposition learning mechanism is implemented to broaden the search scope and augment the variety of the population during the initial phases of iteration. This approach also helps to guarantee the convergence of the algorithm in the later phases of iteration. The unified boundary processing method of delay concentration is improved to solve the problem that a large amount of convergence damages diversity if a large number of transgressive individuals are unified and centralized. The algorithm flow is given and the asymptotic fitness and the asymptotic global optimal solution are proved. The time complexity is analyzed theoretically. Ultimately, six recognized algorithms were utilized to simulate the CEC 2017 function set across different scales. Findings indicate that the enhanced algorithm significantly boosts convergence rate, optimization precision, and solution robustness, demonstrating strong convergence characteristics.

Long-Time Coherent Integration Method for Passive Bistatic Radar Using Frequency Hopping Signals.

Long-time coherent integration using frequency hopping signals is a challenging problem for passive bistatic radar due to its frequency hopping characteristics. Apart from range walk, range curve, and Doppler frequency migration, Doppler diffusion caused by frequency hopping characteristics occurs within the observation time, which also lowers the detection performance. To deal with this problem, a novel coherent integration method for frequency hopping signals based on passive bistatic radar is proposed in this paper. In this novel method, range curve and range walk are eliminated by applying generalized Keystone transform. Then, Doppler frequency migration caused by the target's acceleration is compensated for by a parameter search with a designed search scope. Finally, Doppler frequency migration caused by frequency hopping characteristics is compensated for by designing a new acceleration compensation function and a revised rotation factor for Fourier transform. Since migration effects caused by frequency hopping characteristics are considered and compensated for when using frequency hopping signals, the weak target echo can be better integrated in the observation time compared to when using the existing methods. The simulation results and performance analysis illustrate the effectiveness of the proposed method.

Knowledge-enhanced multidimensional estimation of distribution hyper-heuristic evolutionary algorithm for semiconductor final testing scheduling problem

The semiconductor final test scheduling problem (SFTSP), recognized as a crucial bottleneck in the semiconductor production process, holds immense significance for improving both quality control and scheduling efficiency within chip and integrated circuit enterprises. This article introduces the knowledge-enhanced multidimensional estimation of distribution hyper-heuristic evolutionary algorithm (KMEDHEA) for addressing the SFTSP with the aim of minimizing the makespan. First, a single-vector encoding scheme is used to represent feasible solutions, and a problem-specific constrained-separable left-shift decoding scheme is devised to transform these solutions into feasible scheduling schedules. Second, eight simple yet effective heuristics with problem-specific knowledge are developed that served as a suite of low-level heuristics (LLHs) for exploring the problem solution space. Third, the multidimensional estimation of distribution algorithm (MEDA) is employed as the high-level strategy to estimate the correlations and connections of the pre-designed LLHs, thereby guiding the search scope towards high-quality individuals. Finally, critical configurations of parameters are systematically analyzed by conducting a design-of-experiment (DOE) approach. Numerical experiments are conducted on well-known benchmark datasets, and the experimental results demonstrate the superiority of the KMEDHEA versus several state-of-the-art approaches. The best-known solutions are updated for nine out of ten benchmark instances, highlighting the effectiveness and efficiency of the proposed KMEDHEA in solving the SFTSP.

Enhancing Deep Learning-Based City-Wide Traffic Prediction Pipelines Through Complexity Analysis

Deep learning models can effectively capture the non-linear spatiotemporal dynamics of city-wide traffic forecasting. Evidence of varying deep learning model performance between different cities, different prediction horizons, different scales, specific city regions, and during particular hours of the day abounds in the literature on deep learning-based traffic prediction, yet a unified metric to quantify the complexity of different prediction tasks does not exist. This paper proposes two metrics—model complexity (MC) and intrinsic complexity (IC). While MC quantifies the effective complexity of deep learning models for city-wide traffic prediction tasks, the IC quantifies the underlying complexity of the prediction task. Being an effective complexity metric, MC depends on the model and the data. The IC depends only on the data and is invariant to the model being used. Both metrics are validated through systematic experimentation using traffic volume data from three cities. Finally, we demonstrate how these metrics can improve the workflows for deep learning-based data-driven traffic prediction pipelines and deployment by reducing the hyperparameter search scope and comparing the effectiveness of different design pathways.

Divergent trajectories on frontier innovations: A comparison of international venture capital-invested ventures between China and the United States

How do international venture capital (IVC)-invested ventures (IIVs) between the U.S. and China generate new knowledge differently and follow distinct growth trajectories? Given that venture capital (VC) is critical for financing frontier innovations among new ventures, comparing U.S. and Chinese IIVs is essential for informing national innovation policy. Based on the organizational ecology perspective, we propose that U.S. and Chinese new ventures have divergent trajectories in frontier innovations. We identify 1782 first-round IVC investments in U.S. and Chinese new ventures operating in the computer software, e-commerce, biotechnology, semiconductors, and artificial intelligence sectors between 2010 and 2020 along with their 2825 patents (317 Chinese patents and 2508 U.S. patents). Our findings indicate that Chinese IIVs exhibit greater variation in terms of technological distance, search scope, and knowledge coupling, compared to U.S. IIVs. However, Chinese IIVs’ frontier innovations have more divergent technological trajectories and attract fewer followers than U.S. IIVs, based on patent citations. In a firm-level analysis, we find that Chinese IIVs are less likely to receive a subsequent financing round. Further, IIVs with greater technological distance and search scope, and with higher impact patents, have more opportunities to receive the next round of funding.

Ranking occupations by their proximity to workers’ profiles

Information friction makes it difficult for job seekers to find new employment opportunities. We propose a method for providing individual-specific occupation recommendations by ranking occupations based on their proximity to the worker’s profile. We identify a set of twelve skills, abilities and work styles that capture the worker-oriented requirements of all occupations and discuss how to measure these items using online questions and tasks. We use the Euclidean distance between the measured items pertaining to a worker and the requirements of an occupation to measure the proximity between job seekers and occupations. We show that the proximity between job seekers’ profiles and their preunemployment occupation predicts their intention to change occupations, thus suggesting that our method captures a meaningful conceptualization of mismatch. We also show that our method generates recommendations that differ from the previous occupations of mismatched job seekers, thereby potentially expanding their search scope.

Ethical Considerations in Artificial Intelligence Interventions for Mental Health and Well-Being: Ensuring Responsible Implementation and Impact

AI has the potential to revolutionize mental health services by providing personalized support and improving accessibility. However, it is crucial to address ethical concerns to ensure responsible and beneficial outcomes for individuals. This systematic review examines the ethical considerations surrounding the implementation and impact of artificial intelligence (AI) interventions in the field of mental health and well-being. To ensure a comprehensive analysis, we employed a structured search strategy across top academic databases, including PubMed, PsycINFO, Web of Science, and Scopus. The search scope encompassed articles published from 2014 to 2024, resulting in a review of 51 relevant articles. The review identifies 18 key ethical considerations, including 6 ethical considerations associated with using AI interventions in mental health and wellbeing (privacy and confidentiality, informed consent, bias and fairness, transparency and accountability, autonomy and human agency, and safety and efficacy); 5 ethical principles associated with the development and implementation of AI technologies in mental health settings to ensure responsible practice and positive outcomes (ethical framework, stakeholder engagement, ethical review, bias mitigation, and continuous evaluation and improvement); and 7 practices, guidelines, and recommendations for promoting the ethical use of AI in mental health interventions (adhere to ethical guidelines, ensure transparency, prioritize data privacy and security, mitigate bias and ensure fairness, involve stakeholders, conduct regular ethical reviews, and monitor and evaluate outcomes). This systematic review highlights the importance of ethical considerations in the responsible implementation and impact of AI interventions for mental health and well-being. By addressing privacy, bias, consent, transparency, human oversight, and continuous evaluation, we can ensure that AI interventions like chatbots and AI-enabled medical devices are developed and deployed in an ethically sound manner, respecting individual rights, promoting fairness, and maximizing benefits while minimizing potential harm.

Chaotic RIME optimization algorithm with adaptive mutualism for feature selection problems

The RIME optimization algorithm is a newly developed physics-based optimization algorithm used for solving optimization problems. The RIME algorithm proved high-performing in various fields and domains, providing a high-performance solution. Nevertheless, like many swarm-based optimization algorithms, RIME suffers from many limitations, including the exploration-exploitation balance not being well balanced. In addition, the likelihood of falling into local optimal solutions is high, and the convergence speed still needs some work. Hence, there is room for enhancement in the search mechanism so that various search agents can discover new solutions. The authors suggest an adaptive chaotic version of the RIME algorithm named ACRIME, which incorporates four main improvements, including an intelligent population initialization using chaotic maps, a novel adaptive modified Symbiotic Organism Search (SOS) mutualism phase, a novel mixed mutation strategy, and the utilization of restart strategy. The main goal of these improvements is to improve the variety of the population, achieve a better balance between exploration and exploitation, and improve RIME's local and global search abilities. The study assesses the effectiveness of ACRIME by using the standard benchmark functions of the CEC2005 and CEC2019 benchmarks. The proposed ACRIME is also applied as a feature selection to fourteen various datasets to test its applicability to real-world problems. Besides, the ACRIME algorithm is applied to the COVID-19 classification real problem to test its applicability and performance further. The suggested algorithm is compared to other sophisticated classical and advanced metaheuristics, and its performance is assessed using statistical tests such as Wilcoxon rank-sum and Friedman rank tests. The study demonstrates that ACRIME exhibits a high level of competitiveness and often outperforms competing algorithms. It discovers the optimal subset of features, enhancing the accuracy of classification and minimizing the number of features employed. This study primarily focuses on enhancing the equilibrium between exploration and exploitation, extending the scope of local search.

A Review of Meta-Analyses of Correlation Coefficients on L2 Reading Comprehension

Quantitative meta-analyses have gained popularity among L2 reading researchers in recent years. Meta-analyses, especially meta-analyses that investigate the relationships among variables (e.g., meta-analyses of correlation coefficients, meta-analytic structural equation modeling), are well-suited for L2 reading research due to their componential nature. The present study reviews 14 meta-analyses on L2 reading comprehension published between 2010 and 2023, with a focus on their key study features and findings. While all of the 14 meta-analyses examined the relationship between one or more reading-related variables and L2 reading comprehension and yielded substantial convergence in their findings, their research focus and scope also varied widely, leading to some level of diversity in their scope of literature search, methods of analysis, and findings. We review our findings on the convergence and divergence of the study features across the 14 meta-analyses, summarize their key findings, and provide suggestions for future research.

A Parkinson's Auxiliary Diagnosis Algorithm Based on a Hyperparameter Optimization Method of Deep Learning.

Parkinson's disease is a common mental disease in the world, especially in the middle-aged and elderly groups. Today, clinical diagnosis is the main diagnostic method of Parkinson's disease, but the diagnosis results are not ideal, especially in the early stage of the disease. In this paper, a Parkinson's auxiliary diagnosis algorithm based on a hyperparameter optimization method of deep learning is proposed for the Parkinson's diagnosis. The diagnosis system uses ResNet50 to achieve feature extraction and Parkinson's classification, mainly including speech signal processing part, algorithm improvement part based on Artificial Bee Colony algorithm (ABC) and optimizing the hyperparameters of ResNet50 part. The improved algorithm is called Gbest Dimension Artificial Bee Colony algorithm (GDABC), proposing "Range pruning strategy" which aims at narrowing the scope of search and "Dimension adjustment strategy" which is to adjust gbest dimension by dimension. The accuracy of the diagnosis system in the verification set of Mobile Device Voice Recordings at King's College London (MDVR-CKL) dataset can reach more than 96%. Compared with current Parkinson's sound diagnosis methods and other optimization algorithms, our auxiliary diagnosis system shows better classification performance on the dataset within limited time and resources.

Googling Politics? Comparing Five Computational Methods to Identify Political and News-related Searches from Web Browser Histories

ABSTRACT Search engines play a crucial role in today’s information environment. Yet, political and news-related (PNR) search engine use remains understudied, mainly due to the lack of suitable measurement methods to identify PNR searches. Existing research focuses on specific events, topics, or news articles, neglecting the broader scope of PNR search. Furthermore, self-reporting issues have led researchers to use browsing history data, but scalable methods for analyzing such data are limited. This paper addresses these gaps by comparing five computational methods to identify PNR searches in browsing data, including browsing sequences, context-enhanced dictionary, Traditional Supervised Machine Learning (SML), Transformer-based SML, and zero-shot classification. Using Dutch Google searches as a test case, we use Dutch browsing history data obtained via data donations in May 2022 linked to surveys (N users = 315; N records = 9,868,209; N searches = 697,359), along with 35.5k manually annotated search terms. The findings highlight substantial variation in accuracy, with some methods being more suited for narrower topics. We recommend a two-step approach, applying zero-shot classification followed by human evaluation. This methodology can inform future empirical research on PNR search engine use.