Occurrence Matrix Research Articles

TopicRefiner: Coherence-Guided Steerable LDA for Visual Topic Enhancement.

This article presents a new Human-steerable Topic Modeling (HSTM) technique. Unlike existing techniques commonly relying on matrix decomposition-based topic models, we extend LDA as the fundamental component for extracting topics. LDA's high popularity and technical characteristics, such as better topic quality and no need to cherry-pick terms to construct the document-term matrix, ensure better applicability. Our research revolves around two inherent limitations of LDA. First, the principle of LDA is complex. Its calculation process is stochastic and difficult to control. We thus give a weighting method to incorporate users' refinements into the Gibbs sampling to control LDA. Second, LDA often runs on a corpus with massive terms and documents, forming a vast search space for users to find semantically relevant or irrelevant objects. We thus design a visual editing framework based on the coherence metric, proven to be the most consistent with human perception in assessing topic quality, to guide users' interactive refinements. Cases on two open real-world datasets, participants' performance in a user study, and quantitative experiment results demonstrate the usability and effectiveness of the proposed technique.

Content Classification of the Official Website of the Ministry of Foreign Affairs of the Republic of Indonesia (MoFA RI) using Vector Space Model (VSM)

The official website of the Ministry of Foreign Affairs of the Republic of Indonesia (MoFA RI) is an important platform for disseminating information to a diverse audience. Efficiently categorizing the vast amount of content available on the website is essential for enhancing user experience and optimizing information retrieval. These categories will also become an identifier and topic classification based on the content inside the article. This study presents a systematic approach to content classification of the Official Website of the Ministry of Foreign Affairs of the Republic of Indonesia (MoFA RI) using the Vector Space Model (VSM). The methodology involves preprocessing the text data, constructing a term-document matrix, and implementing cosine similarity to measure the relevance of documents to predefined categories. The study demonstrates the effectiveness of VSM in accurately classifying content, thus facilitating streamlined access to information for users navigating the website. Furthermore, the findings offer insights into enhancing the organization and accessibility of governmental online platforms, contributing to improved user experience and information dissemination.

Biogeography and ecology of the Algerian island flora

Background and aims – In spite of their proximity to the coasts, the flora of the islands and islets of the Algerian coasts has been poorly studied. This work is the first to provide an overall view of the richness and the determinants of the Algerian island flora. Material and methods – The study is based on a compilation of original floristic data. Two classifications of 30 islands and islets were derived from PCA and HCPC performed on the species occurrence matrix and on a matrix of flora descriptors including species richness, functional traits (life form, seed dispersal, pollination), and biogeographic range. We performed an RDA to explain the variation in flora characteristics by a set of physiographic (longitude, latitude, elevation, area, isolation, steepness index, and area/perimeter ratio) and biotic (seabird density and human presence) variables. Key results – The floristic inventory encompassed a total richness of 295 vascular plant taxa (including subspecies and varieties) on the 30 studied sites. Five main groups of islands and islets can be distinguished based on vegetation composition and three from flora descriptors. RDA model selection revealed that the combination of four variables (seabird density, area, latitude, and longitude) explained 26.6% of the variation in flora characteristics. Taken alone, the density of yellow-legged gull colonies and the island area were the main drivers of this variation. Results showed that floristic richness was associated with larger island area and higher seabird density, the importance of some plant functional traits, such as zoochory and entomogamy, and with a higher proportion of Eurasian holarctic species. Conclusion – In the context of island flora conservation, some of these small islands of Algeria can be considered as “modern refuges” from human pressures, and this is particularly important in the context of Mediterranean ecosystems characterised by a quasi-permanent human impact in various habitats. Two new important plant areas (IPAs) for Algeria are proposed following the results of these floristic inventories.

Automatic text classification of drug-induced liver injury using document-term matrix and XGBoost.

Regulatory agencies generate a vast amount of textual data in the review process. For example, drug labeling serves as a valuable resource for regulatory agencies, such as U.S. Food and Drug Administration (FDA) and Europe Medical Agency (EMA), to communicate drug safety and effectiveness information to healthcare professionals and patients. Drug labeling also serves as a resource for pharmacovigilance and drug safety research. Automated text classification would significantly improve the analysis of drug labeling documents and conserve reviewer resources. We utilized artificial intelligence in this study to classify drug-induced liver injury (DILI)-related content from drug labeling documents based on FDA's DILIrank dataset. We employed text mining and XGBoost models and utilized the Preferred Terms of Medical queries for adverse event standards to simplify the elimination of common words and phrases while retaining medical standard terms for FDA and EMA drug label datasets. Then, we constructed a document term matrix using weights computed by Term Frequency-Inverse Document Frequency (TF-IDF) for each included word/term/token. The automatic text classification model exhibited robust performance in predicting DILI, achieving cross-validation AUC scores exceeding 0.90 for both drug labels from FDA and EMA and literature abstracts from the Critical Assessment of Massive Data Analysis (CAMDA). Moreover, the text mining and XGBoost functions demonstrated in this study can be applied to other text processing and classification tasks.

The analysis of association rules: Latent class analysis

AbstractAssociation rules are used to extract information from transactional databases with a collection of items also called “tokens” or “words.” The aim of association rule analysis is to indicate what and how items go with what items in a set of transactions called “documents.” This approach is used in the analysis of text records, of blogs in social media and of shopping baskets. We present here an approach to analyze documents using latent class analysis (LCA) clustering of document term matrices. A document term matrix (DTM) consists of rows referring to documents and columns corresponding to items. In binary weights, “1” indicates the presence of a term in a document and “0” otherwise. The clustering of similar documents provides stratified data sets used to enhance the interpretability of measures of interest such as lift, odds ratios and relative linkage disequilibrium. The article demonstrates the approach with two case studies. A first example consists of comments recorded in a survey aimed at pet owners. A second, much larger example, is based on online reviews to crocs sandals. Association rules describe combinations of terms in the pet survey and crocs reviews. In Section 3, we compute, for these case studies, association rule measures of interest defined in Section 2. We first introduce the case studies to motivate the methods proposed here. In Section 4, we provide a new approach with an enhanced interpretations of measures such as lift by comparing them across clusters derived from an LCA of the DTM. A key result is the application of clustered data in analyzing observational data. This enhances generalizability and interpretability of findings from text analytics. The article concludes with a discussion in Section 5.

A simple and efficient filter feature selection method via document-term matrix unitization

Text processing tasks commonly grapple with the challenge of high dimensionality. One of the most effective solutions to this challenge is to preprocess text data through feature selection methods. Feature selection can select the most advantageous features for subsequent operations (e.g., classification) from the native feature space of the text. This process effectively trims the feature space’s dimensionality, enhancing subsequent operations’ efficiency and accuracy. This paper proposes a straightforward and efficient filter feature selection method based on document-term matrix unitization (DTMU) for text processing. Diverging from previous filter feature selection methods that concentrate on scoring criteria definition, our method achieves more optimal feature selection by unitizing each column of the document-term matrix. This approach mitigates feature-to-feature influence and reinforces the role of the weighting proportion within the features. Subsequently, our scoring criterion subtracts the sum of weights for negative samples from positive samples and takes the absolute value. We conduct numerical experiments to compare DTMU with four advanced filter feature selection methods: max–min ratio metric, proportional rough feature selector, least loss, and relative discrimination criterion, along with two classical filter feature selection methods: Chi-square and information gain. The experiments are performed on four ten-thousand-dimensional feature space datasets: book, dvd, music, movie and two thousand-dimensional feature space datasets: imdb, amazon_cells, sourced from Amazon product reviews and movie reviews. Experimental findings demonstrate that DTMU selects more advantageous features for subsequent operations and achieves a higher dimensionality reduction rate than those of the other six methods used for comparison. Moreover, DTMU exhibits robust generalization capabilities across various classifiers and dimensional datasets. Notably, the average CPU time for a single run of DTMU is measured at 1.455 s.

Residual U-Net approach for thyroid nodule detection and classification from thyroid ultrasound images

With so many thyroid knobs (nodules) discovered by accident, it is critical to recognize as many aberrant knobs (nodules) as possible from fine-needle aspiration (FNA) biopsies or other medical procedures while excluding those that are virtually certainly benign. Thyroid ultrasonography, on the other hand, is prone to interobserver variability and subjective translations. An effective deep learning model for segmenting and categorizing thyroid nodules in this study follows the stages below: data collection from a well-known archive, The Thyroid Digital Image Database (TDID), which comprises ultrasound pictures from 298 patients, preprocessing using anisotropic diffusion filter (ADF) for removing noise and enhancing the images, segmentation using a bilateral filter for segmenting images, feature extraction using grey level occurrence matrix (GLCM), feature selection using Multi-objective Particle Swarm with Random Forest Optimization (MbPSRA) and finally classification happens were Residual U-Net will be used. Experiment evaluation states the proposed model outperforms well than other state-of-art models.

A Large-Scale Sensitivity Analysis on Latent Embeddings and Dimensionality Reductions for Text Spatializations.

The semantic similarity between documents of a text corpus can be visualized using map-like metaphors based on twodimensional scatterplot layouts. These layouts result from a dimensionality reduction on the document-term matrix or a representation within a latent embedding, including topic models. Thereby, the resulting layout depends on the input data and hyperparameters of the dimensionality reduction and is therefore affected by changes in them. Furthermore, the resulting layout is affected by changes in the input data and hyperparameters of the dimensionality reduction. However, such changes to the layout require additional cognitive efforts from the user. In this work, we present a sensitivity study that analyzes the stability of these layouts concerning (1) changes in the text corpora, (2) changes in the hyperparameter, and (3) randomness in the initialization. Our approach has two stages: data measurement and data analysis. First, we derived layouts for the combination of three text corpora and six text embeddings and a grid-search-inspired hyperparameter selection of the dimensionality reductions. Afterward, we quantified the similarity of the layouts through ten metrics, concerning local and global structures and class separation. Second, we analyzed the resulting 42 817 tabular data points in a descriptive statistical analysis. From this, we derived guidelines for informed decisions on the layout algorithm and highlight specific hyperparameter settings. We provide our implementation as a Git repository at hpicgs/Topic-Models-and-DimensionalityReduction-Sensitivity-Study and results as Zenodo archive at DOI:10.5281/zenodo.12772898.

Stylistic Conversion and Linguistic Beauty Reproduction in English Translation-Based on Neurosemantic Analysis

Abstract The core of translation is language conversion, but it should be more of an aesthetic experience, a transformation of different ways of thinking. This paper proposes a neural semantic analysis method for English translation based on text similarity calculation. The technique uses a large-scale corpus to train the vector space model of words, establishes a concept-context matrix through the term-document matrix, and determines each value of the concept vector with a weight according to the literary translation in the knowledge base. The information extraction and evaluation of the English-translated text through semantic similarity calculation provides a theoretical basis for the subsequent analysis of style conversion and linguistic, and aesthetic reproduction of the English-translated text. The study will take the Lin Yutang translation and the Ricardian translation of The Analects of Confucius as an example to develop the analysis of style transformation and reproduction of linguistic beauty of the English-translated text from both lexical and syntactic levels. The results show that the lexical density of Lin Yutang’s translation is 0.5866, while that of Rie Jacob’s translation is 0.5334, with the former significantly larger than the latter. It shows that Lin Yutang’s translation of The Analects of Confucius adopts a naturalization strategy, i.e., To disseminate Chinese culture in a way that is authentic, familiar to English readers, and easy to understand, so as to make his linguistic expression more emotional.

In search of generality: Revised distribution data and regionalization of Cerrado endemic tetrapods

AbstractAimTo search for a general regionalization pattern using verified records of endemic terrestrial vertebrates. To test previous hypotheses of congruent distribution patterns for Cerrado biotas. To study the role of elevation as a driver of endemism and distribution in the Cerrado region.LocationCerrado domain, central South America.TaxonTetrapoda.MethodsWe used a species occurrence matrix to implement a biotic element (BE) analysis to test for non‐random spatial congruence of raw distributions of Cerrado endemic amphibians, reptiles, birds and mammals. We implemented a Kruskal–Wallis test on elevational records to classify species as occurring predominantly in plateaus or depressions, and to classify all delimited BEs in such categories.ResultsWe compiled and revised 13,800 unique distribution records of 340 Cerrado endemic tetrapods and detected a significant, non‐random co‐distribution pattern formed by 29 BEscomprising 182 species, corroborating the first general prediction of the vicariant model. Most BEs were composed of at least three vertebrate groups, reflecting general tetrapod endemism patterns. Congeneric species were segregated among different BEs, corroborating the second prediction of the vicariant model. Our regionalization scheme was broadly congruent with previous results and revealed three previously undetected areas. Most (89%) partial or restricted BEs are clearly segregated in elevation, and we recognized 14 ‘Plateau units’ and 12 ‘Depression units’.Main ConclusionsOur results support an emergent consensual biotic regionalization pattern in the Cerrado. We also revealed three novel regions and complex biogeographical patterns. The clear altitudinal segregation among BEs validates previous hypotheses on the role of geomorphological changes in shaping Cerrado and Neotropical endemism patterns. Our recovered units should serve as a template for the description of new taxa, for delimiting poorly studied ranges, and for guiding urgent conservation action in the richest and most imperilled tropical savanna in the globe.

Sentiment Analysis of the Top 5 E-commerce Platforms in Indonesia using Text Mining and Natural Language Processing (NLP)

This research attempts to depict a sentiment comparison of the top 5 E-commerce platforms in Indonesia by gathering the emotional tone behind sentence contents related to customer sentiments, customer experiences, and the brand reputation of E-commerce. Data were collected using Python 3.11.4 with the google-play-scraper library, extracted from user reviews/comments on each play store page of the top 5 E-commerce platforms in Indonesia. A sampling of 10,000 records was taken to form a long document term matrix (DTM) of 59,981,785 due to the limitation of CPU capacity for data matrix size. R Programming version 4.3.1 was employed for sentiment analysis in this study. It can be concluded that user comments or reviews on the top five (5) E-commerce platforms in Indonesia show positive sentences indicating user satisfaction (3664 sentences), neutral sentences indicating average user appreciation (2282 sentences), and negative sentences indicating user dissatisfaction (4054 sentences). At least with more positive and neutral sentences, it is indicated that 59.64% of E-commerce users in Indonesia express a positive opinion on the performance of the top 5 E-commerce platforms in the country.

Normalized effect size (NES): a novel feature selection model for Urdu fake news classification

Social media has become an essential source of news for everyday users. However, the rise of fake news on social media has made it more difficult for users to trust the information on these platforms. Most research studies focus on fake news detection in the English language, and only a limited number of studies deal with fake news in resource-poor languages such as Urdu. This article proposes a globally weighted term selection approach named normalized effect size (NES) to select highly discriminative features for Urdu fake news classification. The proposed model is based on the traditional inverse document frequency (TF-IDF) weighting measure. TF-IDF transforms the textual data into a weighted term-document matrix and is usually prone to the curse of dimensionality. Our novel statistical model filters the most discriminative terms to reduce the data’s dimensionality and improve classification accuracy. We compare the proposed approach with the seven well-known feature selection and ranking techniques, namely normalized difference measure (NDM), bi-normal separation (BNS), odds ratio (OR), GINI, distinguished feature selector (DFS), information gain (IG), and Chi square (Chi). Our ensemble-based approach achieves high performance on two benchmark datasets, BET and UFN, achieving an accuracy of 88% and 90%, respectively.

A Framework for the Unsupervised and Semi-Supervised Analysis of Visual Frames

AbstractThis article introduces to political science a framework to analyze the content of visual material through unsupervised and semi-supervised methods. It details the implementation of a tool from the computer vision field, the Bag of Visual Words (BoVW), for the definition and extraction of “tokens” that allow researchers to build an Image-Visual Word Matrix which emulates the Document-Term matrix in text analysis. This reduction technique is the basis for several tools familiar to social scientists, such as topic models, that permit exploratory, and semi-supervised analysis of images. The framework has gains in transparency, interpretability, and inclusion of domain knowledge with respect to other deep learning techniques. I illustrate the scope of the BoVW by conducting a novel visual structural topic model which focuses substantively on the identification of visual frames from the pictures of the migrant caravan from Central America.

Sintigrafik Görüntülerden Tiroid Nodülleri için Bilgisayar Destekli Tanı Sistemi

In modern medicine, image recognition via segmentation of anatomical regions and automatic classification of diseases using medical images has a growing potential role in diagnosis of various diseases. Scintigraphy of thyroid is one of the established imaging modalities for diagnosis of thyroid gland disorders. In our study, the speckle noise was reduced in the scintigraphy images with the optimized Bayesian nonlocal mean filter. The thyroid gland was automatically segmented by local based active contour method and the thyroid gland pathologies were classified with convolutional neural networks (CNN). The proposed computer aided diagnosis (CAD) system was compared with Pyramid of Histograms of Orientation Gradients (PHOG), Gray Level Co occurrence Matrix (GLCM), Local Configuration Pattern (LCP) and Bag of Feature (BoF) methods. The common pathological patterns of scintigraphic images of the thyroid gland were successfully classified by CNN with an overall success rate of 91.19%. The comparative methods were PHOG, GLCM, LCP and BoF methods which provided overall success rates of 7.61%, 86.04%, 88.91% and 85.72% respectively. The proposed CNN based automatic diagnosis system provided promising results compared to handcrafted methods.

Improving information retrieval through correspondence analysis instead of latent semantic analysis

The initial dimensions extracted by latent semantic analysis (LSA) of a document-term matrix have been shown to mainly display marginal effects, which are irrelevant for information retrieval. To improve the performance of LSA, usually the elements of the raw document-term matrix are weighted and the weighting exponent of singular values can be adjusted. An alternative information retrieval technique that ignores the marginal effects is correspondence analysis (CA). In this paper, the information retrieval performance of LSA and CA is empirically compared. Moreover, it is explored whether the two weightings also improve the performance of CA. The results for four empirical datasets show that CA always performs better than LSA. Weighting the elements of the raw data matrix can improve CA; however, it is data dependent and the improvement is small. Adjusting the singular value weighting exponent often improves the performance of CA; however, the extent of the improvement depends on the dataset and the number of dimensions.

Geochemical control, water quality indexing, source distribution, and potential health risk of fluoride and arsenic in groundwater: Occurrence, sources apportionment, and positive matrix factorization model

Fluoride (F-), and arsenic (As) in the groundwater cause health problems in developing countries, including Pakistan. We evaluated the occurrence, distribution, sources apportionment, and health hazards of F-, and As in the groundwater of Mardan, Pakistan. Therefore, groundwater samples (n = 130) were collected and then analyzed for F-, and As by ion-chromatography (IC) and Inductively-coupled plasma mass-spectrometry (ICP-MS). The F-, and As concentrations in groundwater were 0.7–14.4 mg/L and 0.5–11.2 µg/L. Relatively elevated F-, and As coexists with higher pH, Na+, HCO3-, SO4−2, and depleted Ca+2 due to fluoride, sulfide-bearing minerals, and anthropogenic inputs. Both F-, and/or As are transported in subsurface water through adsorption and desorption processes. Groundwater samples 45%, and 14.2% exceeded the WHO guidelines of 1.5 mg/L and 10 µg/L. Water quality indexing (WQI-model) declared that 35.7% samples are unfit for household purposes. Saturation and undersaturation of minerals showed precipitation and mineral dissolution. Groundwater contamination by PCA-MLR and PMF-model interpreted five factors. The fitting results and R2 values of PMF (0.52–0.99)>PCA-MLR (0.50–0.95) showed high accuracy of PMF-model. Human health risk assessment (HHRA-model) revealed high non-carcinogenic and carcinogenic risk for children than adults. The percentile recovery of F- and As was recorded 98%, and 95% with reproducibility ± 5% error.

ALLERGIC CONTACT DERMATITIS: ETIOLOGICAL MATRIX

The article discusses the main links of the pathogenesis of allergic contact dermatitis and gives the results of patch testing in patients who were examined to determine the etiological factors of the disease. Screening showed that nickel, cobalt, methylisothiazolinone, phenylenediamine, formaldehydes and fragrances are the most common etiological factors of ACD. Named haptens cause the absolute majority of positive reactions in the selected group of patients, and also cause the largest number of reactions of pronounced intensity. Patch testing is an informative and useful method for a balanced planning in ACD patients treatment strategy, because it reveals the etiological matrix of the pathological process occurrence in the skin.

A new approach for determining optimum number of cell, cell formation and intra-machine cell layout with considering operation sequence

Cellular manufacturing has been effective in implementing Group Technology Philosophy so as to design a manufacturing system. In cellular manufacturing, it is imperative for the machines to be grouped and the parts to be assigned as part families depending on the level of similarity. Several research works have been carried out on the challenges faced in cell formation and their variants. Yet, based on zero-one machine-product occurrence matrix as the study's contribution data, only a small number of researchers have come out with a solution for overcoming feasibility evaluation and cell arrangement problems. An innovative similarity coefficient approach was proposed in the study to integrate feasibility evaluation, cell creation and intra-machine cell layout design by taking operation sequence into account. For tackling the challenges, an efficient heuristic solution has been proposed to use the similarity coefficient. The purpose of this study is to use eigenvalues of the resemblance matrix of coefficient and Kaiser's law to determine the appropriate sum of cells in the machine-product occurrence matrix. It is followed by rearranging the machines as well as the product depending on similarity value. The intra-machine cell layout design depending on flow matrix is attained. In addition, the proposed approach takes into consideration the intercellular movements, backtracking movements, number of operation and voids. The performance of the proposed approach has been assessed with renowned bench mark problem derived from previous literature and the findings have been contrasted to the CASE and CLASS algorithm. From the computational result, it could be inferred that the suggested approach is found to offer better (or) equal solution than the prevailing method. To add more, a real-world industrial case has been suggested to make evident the manner in which the proposed approach functions and the benefits attained by implementing the same.

The Children's Picture Books Lexicon (CPB-LEX): A large-scale lexical database from children's picture books.

This article presents CPB-LEX, a large-scale database of lexical statistics derived from children's picture books (age range 0-8 years). Such a database is essential for research in psychology, education and computational modelling, where rich details on the vocabulary of early print exposure are required. CPB-LEX was built through an innovative method of computationally extracting lexical information from automatic speech-to-text captions and subtitle tracks generated from social media channels dedicated to reading picture books aloud. It consists of approximately 25,585 types (wordforms) and their frequency norms (raw and Zipf-transformed), a lexicon of bigrams (two-word sequences and their transitional probabilities) and a document-term matrix (which shows the importance of each word in the corpus in each book). Several immediate contributions of CPB-LEX to behavioural science research are reported, including that the new CPB-LEX frequency norms strongly predict age of acquisition and outperform comparable child-input lexical databases. The database allows researchers and practitioners to extract lexical statistics for high-frequency words which can be used to develop word lists. The paper concludes with an investigation of how CPB-LEX can be used to extend recent modelling research on the lexical diversity children receive from picture books in addition to child-directed speech. Our model shows that the vocabulary input from a relatively small number of picture books can dramatically enrich vocabulary exposure from child-directed speech and potentially assist children with vocabulary input deficits. The database is freely available from the Open Science Framework repository: https://tinyurl.com/4este73c .

Pediatric Injury Surveillance From Uncoded Emergency Department Admission Records in Italy: Machine Learning-Based Text-Mining Approach.

Unintentional injury is the leading cause of death in young children. Emergency department (ED) diagnoses are a useful source of information for injury epidemiological surveillance purposes. However, ED data collection systems often use free-text fields to report patient diagnoses. Machine learning techniques (MLTs) are powerful tools for automatic text classification. The MLT system is useful to improve injury surveillance by speeding up the manual free-text coding tasks of ED diagnoses. This research aims to develop a tool for automatic free-text classification of ED diagnoses to automatically identify injury cases. The automatic classification system also serves for epidemiological purposes to identify the burden of pediatric injuries in Padua, a large province in the Veneto region in the Northeast Italy. The study includes 283,468 pediatric admissions between 2007 and 2018 to the Padova University Hospital ED, a large referral center in Northern Italy. Each record reports a diagnosis by free text. The records are standard tools for reporting patient diagnoses. An expert pediatrician manually classified a randomly extracted sample of approximately 40,000 diagnoses. This study sample served as the gold standard to train an MLT classifier. After preprocessing, a document-term matrix was created. The machine learning classifiers, including decision tree, random forest, gradient boosting method (GBM), and support vector machine (SVM), were tuned by 4-fold cross-validation. The injury diagnoses were classified into 3 hierarchical classification tasks, as follows: injury versus noninjury (task A), intentional versus unintentional injury (task B), and type of unintentional injury (task C), according to the World Health Organization classification of injuries. The SVM classifier achieved the highest performance accuracy (94.14%) in classifying injury versus noninjury cases (task A). The GBM method produced the best results (92% accuracy) for the unintentional and intentional injury classification task (task B). The highest accuracy for the unintentional injury subclassification (task C) was achieved by the SVM classifier. The SVM, random forest, and GBM algorithms performed similarly against the gold standard across different tasks. This study shows that MLTs are promising techniques for improving epidemiological surveillance, allowing for the automatic classification of pediatric ED free-text diagnoses. The MLTs revealed a suitable classification performance, especially for general injuries and intentional injury classification. This automatic classification could facilitate the epidemiological surveillance of pediatric injuries by also reducing the health professionals' efforts in manually classifying diagnoses for research purposes.