Taxonomy Learning Research Articles

Adaptive Taxonomy Learning and Historical Patterns Modeling for Patent Classification

Patent classification aims to assign multiple International Patent Classification (IPC) codes to a given patent. Existing methods for automated patent classification primarily focus on analyzing the text descriptions of patents. However, apart from the textual information, each patent is also associated with some assignees, and the knowledge of their previously applied patents can often be valuable for accurate classification. Furthermore, the hierarchical taxonomy defined by the IPC system provides crucial contextual information and enables models to leverage the correlations between IPC codes for improved classification accuracy. However, existing methods fail to incorporate the above aspects and lead to reduced performance. To address these limitations, we propose an integrated framework that comprehensively considers patent-related information for patent classification. To be specific, we first present an IPC codes correlations learning module to capture both horizontal and vertical information within the IPC codes. This module effectively captures the correlations by adaptively exchanging and aggregating messages among IPC codes at the same level (horizontal information) and from both parent and children codes (vertical information), which allows for a comprehensive integration of knowledge and relationships within the IPC hierarchical taxonomy. Additionally, we design a historical application patterns learning component to incorporate previous patents of the corresponding assignee by aggregating high-order temporal information via a dual-channel graph neural network. Finally, our approach combines the contextual information from patent texts, which encompasses the semantics of IPC codes, with assignees’ sequential preferences to make predictions. Experimental evaluations on real-world datasets demonstrate the superiority of our proposed approach over existing methods. Moreover, we present the model’s ability to capture the temporal patterns of assignees and the semantic dependencies among IPC codes.

Which is better? Taxonomy induction with learning the optimal structure via contrastive learning

A taxonomy represents a hierarchically structured knowledge graph that forms the infrastructure for various downstream applications, including recommender systems, web search, and question answering. The exploration of automated induction from text corpora has yielded notable taxonomies such as CN-probase, CN-DBpedia, and Zhishi.schema. Despite these efforts, existing taxonomies still face two critical issues that result in sub-optimal hierarchical structures. On the one hand, commonly observed taxonomies exhibit a coarse-grained and “flat” structure, stemming from a noticeable lack of diversity in both nodes and edges. This limitation primarily originates from the biased and homogeneous data distribution. On the other hand, the semantic granularity among “siblings” within these taxonomies remains inconsistent, presenting a challenge in accurately and comprehensively identifying hierarchical relations. To address these issues, this study introduces a novel taxonomy induction framework composed of three meticulously designed components. Initially, we established a seed schema by leveraging statistical information from external data sources as distant supervision to append nodes and edges containing “generic semantics”, thereby rectifying biased data distributions. Subsequently, a clustering algorithm is employed to group the nodes based on their similarities, followed by a refinement operation of the hierarchical relations among these nodes. Building on this seed schema, we propose a fine-grained contrastive learning method in the expansion module to strengthen the utilization of taxonomic structures, consequently boosting the precision of query-anchor matching. Finally, we meticulously scrutinized the hierarchical relations between each query and its siblings to ensure the integrity of the constructed taxonomy. Extensive experiments on real-world datasets validated the efficacy of our proposed framework for constructing well-structured taxonomies.

Knowledge-guided generative artificial intelligence for automated taxonomy learning from drug labels.

To automatically construct a drug indication taxonomy from drug labels using generative Artificial Intelligence (AI) represented by the Large Language Model (LLM) GPT-4 and real-world evidence (RWE). We extracted indication terms from 46421 free-text drug labels using GPT-4, iteratively and recursively generated indication concepts and inferred indication concept-to-concept and concept-to-term subsumption relations by integrating GPT-4 with RWE, and created a drug indication taxonomy. Quantitative and qualitative evaluations involving domain experts were performed for cardiovascular (CVD), Endocrine, and Genitourinary system diseases. 2909 drug indication terms were extracted and assigned into 24 high-level indication categories (ie, initially generated concepts), each of which was expanded into a sub-taxonomy. For example, the CVD sub-taxonomy contains 242 concepts, spanning a depth of 11, with 170 being leaf nodes. It collectively covers a total of 234 indication terms associated with 189 distinct drugs. The accuracies of GPT-4 on determining the drug indication hierarchy exceeded 0.7 with "good to very good" inter-rater reliability. However, the accuracies of the concept-to-term subsumption relation checking varied greatly, with "fair to moderate" reliability. We successfully used generative AI and RWE to create a taxonomy, with drug indications adequately consistent with domain expert expectations. We show that LLMs are good at deriving their own concept hierarchies but still fall short in determining the subsumption relations between concepts and terms in unregulated language from free-text drug labels, which is the same hard task for human experts.

Genomic fingerprints of the world's soil ecosystems.

Despite the explosion of soil metagenomic data, we lack a synthesized understanding of patterns in the distribution and functions of soil microorganisms. These patterns are critical to predictions of soil microbiome responses to climate change and resulting feedbacks that regulate greenhouse gas release from soils. To address this gap, we assay 1,512 manually curated soil metagenomes using complementary annotation databases, read-based taxonomy, and machine learning to extract multidimensional genomic fingerprints of global soil microbiomes. Our objective is to uncover novel biogeographical patterns of soil microbiomes across environmental factors and ecological biomes with high molecular resolution. We reveal shifts in the potential for (i) microbial nutrient acquisition across pH gradients; (ii) stress-, transport-, and redox-based processes across changes in soil bulk density; and (iii) greenhouse gas emissions across biomes. We also use an unsupervised approach to reveal a collection of soils with distinct genomic signatures, characterized by coordinated changes in soil organic carbon, nitrogen, and cation exchange capacity and in bulk density and clay content that may ultimately reflect soil environments with high microbial activity. Genomic fingerprints for these soils highlight the importance of resource scavenging, plant-microbe interactions, fungi, and heterotrophic metabolisms. Across all analyses, we observed phylogenetic coherence in soil microbiomes-more closely related microorganisms tended to move congruently in response to soil factors. Collectively, the genomic fingerprints uncovered here present a basis for global patterns in the microbial mechanisms underlying soil biogeochemistry and help beget tractable microbial reaction networks for incorporation into process-based models of soil carbon and nutrient cycling.IMPORTANCEWe address a critical gap in our understanding of soil microorganisms and their functions, which have a profound impact on our environment. We analyzed 1,512 global soils with advanced analytics to create detailed genetic profiles (fingerprints) of soil microbiomes. Our work reveals novel patterns in how microorganisms are distributed across different soil environments. For instance, we discovered shifts in microbial potential to acquire nutrients in relation to soil acidity, as well as changes in stress responses and potential greenhouse gas emissions linked to soil structure. We also identified soils with putative high activity that had unique genomic characteristics surrounding resource acquisition, plant-microbe interactions, and fungal activity. Finally, we observed that closely related microorganisms tend to respond in similar ways to changes in their surroundings. Our work is a significant step toward comprehending the intricate world of soil microorganisms and its role in the global climate.

Advancing the discourse: A next-generation value chain-based taxonomy for circular economy key performance indicators

The growth of interest in circular economy (CE) has been accompanied by different approaches to measurement of CE outcomes and impacts, leading to a wide portfolio of indicators with varying degrees of overlap, inconsistency, and convergence. The aim of this paper is to propose a unifying framework for CE indicators, as the next generation of CE taxonomies. We first undertake a scoping review of 59 review papers on CE indicators using manual and computational methods (i.e., topic modelling) to inform the taxonomy structure and content. As a result, we report on 11 clusters of approaches that have been attentive to different dimensions of CE (e.g. horizontal value chain, vertical scale of operation (macro, meso, micro), impact category (economic, biophysical, social), material vs product focus, etc.). Highlighting the strengths and weakness of these approaches, we identify gaps in dimensions related to horizontal and vertical scales of measurement, and propose an agnostic, integrative framework that builds on the scientific foundations of previous research, within a more systemic and comprehensive taxonomy. This taxonomy could be used as a guiding framework or heuristic for regulators, both nationally and internationally, and for practitioners to undertake a comprehensive measurement and assessment of CE related interventions and initiatives at scale.

Comparing the performance of artificial intelligence learning models to medical students in solving histology and embryology multiple choice questions

IntroductionThe appearance of artificial intelligence language models (AI LMs) in the form of chatbots has gained a lot of popularity worldwide, potentially interfering with different aspects of education, including medical education as well. The present study aims to assess the accuracy and consistency of different AI LMs regarding the histology and embryology knowledge obtained during the 1st year of medical studies. MethodsFive different chatbots (ChatGPT, Bing AI, Bard AI, Perplexity AI, and ChatSonic) were given two sets of multiple-choice questions (MCQs). AI LMs test results were compared to the same test results obtained from 1st year medical students. Chatbots were instructed to use revised Bloom’s taxonomy when classifying questions depending on hierarchical cognitive domains. Simultaneously, two histology teachers independently rated the questions applying the same criteria, followed by the comparison between chatbots’ and teachers’ question classification. The consistency of chatbots’ answers was explored by giving the chatbots the same tests two months apart. ResultsAI LMs successfully and correctly solved MCQs regarding histology and embryology material. All five chatbots showed better results than the 1st year medical students on both histology and embryology tests. Chatbots showed poor results when asked to classify the questions according to revised Bloom’s cognitive taxonomy compared to teachers. There was an inverse correlation between the difficulty of questions and their correct classification by the chatbots. Retesting the chatbots after two months showed a lack of consistency concerning both MCQs answers and question classification according to revised Bloom’s taxonomy learning stage. ConclusionDespite the ability of certain chatbots to provide correct answers to the majority of diverse and heterogeneous questions, a lack of consistency in answers over time warrants their careful use as a medical education tool.

Systematization of autonomous vehicles in last mile transportation processes – taxonomy development and clustering of existing concepts

ABSTRACT With the increasing challenges in last mile logistics, a variety of promising approaches of autonomous delivery concepts have been emerging in recent years. Due to the fast pace of advancements, several approaches have been developed, the characterization of them often by purpose rather than systematically. This study builds upon a multi-stage research process that includes a systematic literature review and structured taxonomy building process, as well as a hierarchical cluster analysis of 45 existing autonomous delivery concepts. Based on that, two main results were achieved. First, a holistic taxonomy with 13 categories - including technological, process-related and customer acceptance characteristics - was developed to achieve characterizing of autonomous delivery concepts in a standardized way. The second result centered on the taxonomy, where four clusters could be identified through the hierarchical cluster analysis. With the developed clusters it is possible to better categorize and compare individual autonomous delivery concepts.

An analysis of discussions in collaborative knowledge engineering through the lens of Wikidata

We study discussions in Wikidata, the world’s largest open-source collaborative knowledge graph (KG). This is important because it helps KG community managers understand how discussions are used and inform the design of collaborative practices and support tools. We follow a mixed-methods approach with descriptive statistics, thematic analysis, and statistical tests to investigate how much discussions in Wikidata are used, what they are used for, and how they support knowledge engineering (KE) activities. The study covers three core sources of discussion, the talk pages that accompany Wikidata items and properties, and a general-purpose communication page. Our findings show low use of discussion capabilities and a power-law distribution similar to other KE projects such as Schema.org. When discussions are used, they are mostly about KE activities, including activities that span across the entire KE lifecycle from conceptualisation and implementation to maintenance and taxonomy building. We hope that the findings will help Wikidata devise improved practices and capabilities to encourage the use of discussions as a tool to collaborate, improve editor engagement, and engineer better KGs.

Exploring design elements of personal data markets

Since the emerging information economy relies heavily on data for advancement and growth, data markets have gained increasing attention. However, while global data economies are evolving and data are increasingly shared among organizations in various data ecosystems, marketplaces for personal data (PDMs) exhibited considerable start-up difficulties, which doomed their majority to either fail quickly or to operate in legal grey zones. Apparently, in recent times, novel PDMs have arisen which seem economically and technically viable. The study investigates this “new generation” from both an economic and a technological perspective. Adhering to a rigorous methodology for taxonomy building and evaluation, 18 dimensions and 59 characteristics are presented alongside which these new PDMs can be designed. Additionally, archetypes are derived. The findings reveal that PDMs tend to follow certain design commonalities holding for data markets generally but comprise specific design elements distinguishing them both from conventional data markets and among each other.

Evaluasi Pembelajaran Pendidikan Agama Islam dan Budi Pekerti Di SMP Negeri

This research was made to find out the process of implementing the evaluation of Islamic religious education and morals in junior high school 1 Ngariboyo Magetan through the interview method. The theoretical basis used is based on the theory of learning evaluation, such as Bloom's taxonomy and other learning evaluation theories. Things discussed around the evaluation of Islamic religious education and morals include preparation of learning evaluations, assessment lists, daily grades, and written and oral exams. The obstacle in undergoing a learning evaluation is the limited teaching staff for Islamic religious education at this school, so the teaching staff needs to use a more effective and efficient way of conducting learning evaluations consisting of 21 class groups, with details for each class level there are 7 classes, there are also other obstacles that are obtained, namely the uneven understanding of students so that in preparing learning evaluations it is necessary to evaluate learning according to student’s abilities. Whereas the assessment standards are made according to the average ability of students, for students who have not been able to meet the assessment standards, a reassessment will be carried out.

Automatic generation of incremental taxonomies for supporting the users in the development of an RPA project

The robotic process automation (RPA) paradigm is a discipline that is becoming increasingly popular thanks to the great interest shown by the industry. In such context, RPA solutions based on artificial intelligence, i.e., cognitive solutions, are receiving increasing attention. In a cognitive RPA project, the RPA developer is in charge of selecting the most suitable components that solve specific tasks from the sets of components provided by different RPA platforms. This selection is very challenging, especially since there is no homogeneity in component names or component classifications. Such a situation turns an RPA project’s development into a time-consuming, error-prone, and very tedious process. Therefore, supporting the RPA developer in developing a cognitive RPA project is desired. The industry has also pointed out this need. This work presents a proposal for supporting the users in developing a cognitive RPA project. To be more precise, an incremental method to automatically generate taxonomies from cognitive RPA platforms is proposed. Such taxonomies can be dynamically adapted when necessary. In previous work, the initial aspects of this research were presented. However, the current work greatly enhances such previous work by: (1) extending the proposed method to improve the management of real-world use cases from industry, (2) developing a proof-of-concept tool that is based on the proposed approach, (3) validating the proposed method by applying it to real-world use cases from industry, and (4) performing a literature review on related topics. The results obtained are auspicious and demonstrate that the proposed approach substantially improves the support given to users during the development of a cognitive RPA project.

Towards a taxonomy of research areas in open government data

PurposeThis study aimed to develop a taxonomy of research areas in open government data (OGD) through a bibliometric mapping tool and a qualitative analysis software.Design/methodology/approachIn this study, the authors extracted metadata of 442 documents from a bibliographic database. The authors used a bibliometric mapping tool for familiarization with the literature. After that, the authors used qualitative analysis software to develop taxonomy.FindingsThis paper developed taxonomy of OGD with three research areas: implementation and management, architecture, users and utilization. These research areas are further analyzed into seven topics and twenty-eight subtopics. The present study extends Charalabidis et al. (2016) taxonomy by adding two research topics, namely the adoption factors and barriers of OGD implementations and OGD ecosystems. Also, the authors include artificial intelligence in the taxonomy as an emerging research interest in the literature. The authors suggest four directions for future research: indigenous knowledge in open data, open data at local governments, development of OGD-specific theories and user studies in certain research themes.Practical implicationsEarly career researchers and doctoral students can use the taxonomy to familiarize themselves with the literature. Also, established researchers can use the proposed taxonomy to inform future research. Taxonomy-building procedures in this study are applicable to other fields.Originality/valueThis study developed a novel taxonomy of research areas in OGD. Taxonomy building is significant because there is insufficient taxonomy of research areas in this discipline. Also, conceptual knowledge through taxonomy creation is a basis for theorizing and theory-building for future studies.

Utilizing The iNaturalist Application for Biology Research and Learning

This literature study aims to describe the utilization of the iNaturalist application (https://www.inaturalist.org/) in biology research and learning activities. iNaturalist is an Online Citizen Science (OCS) application available on the web and Android (Apps) platforms. This application has various features for collecting, identifying, and mapping varied biodiversity worldwide. It is suitable for biodiversity research and can also be used in learning biology on relevant concepts such as biodiversity, ecology, environment, plantae, animalia, and others. This article will discuss the advantages of the iNaturalist application, how to use it, and the utilization of the application in learning biology. The working principle of the iNaturalist application is to utilize artificial intelligence technology in the form of image processing so that it can identify various uploaded biodiversity data. This application offers community learning, the process of identifying research data involving the iNaturalist community worldwide. The way the iNaturalist application works is that users record data, share data with fellow naturalists and discuss research results. iNaturalist in research can be used for collecting, identifying, and publishing ecological, environmental, and biodiversity research data. The use of iNaturalist application in learning can be used in biology learning to train science process skills, especially observation skills, biodiversity learning, taxonomy learning, and research skills. In addition, this application supports Citizen Science activities that are suitable for use in education biology because it can create contextual and collaborative learning.

A framework to design game theory-based interventions for strategic analysis of real-world problems with stakeholders

Game theory provides a valuable conceptual framework for developing strategic insight into complex, multi-actor situations and is well-suited for addressing real-world problems. Participatory game theory involves model construction and analysis with stakeholders guided by a facilitator with the aim of examining strategic interactions within real-world problems. Previous work in this field focused often on the analytical insights from game theory models whereas guidance on the design and implementation process of participatory game theory interventions remains under researched. This paper addresses this gap by providing a framework to assist in the design of game theory-based interventions. Here, stakeholders’ decision-support needs are designed according to specific learning levels, as defined by Bloom's taxonomy and triple-loop learning, that the intervention will target. The framework further specifies the design choices for each learning level by comparing the design choices from two case studies of game theory-based interventions in Dutch railway and Bangladesh drinking water contexts. The former involved stakeholders using a game concept identification tool while the latter used game theory-based role-playing games with community and local government stakeholders. This paper describes the value of this framework for structured learning through game theory-based interventions. It includes the facilitator's role and design choices for each learning level related to the knowledge capabilities and activity focus to achieve that learning outcome. Future work will incorporate design inputs from additional learning theories, behavioural game theory and will explore the framework's potential for evaluating game theory-based interventions.

Development of MIKiR Teaching Materials Based on Educational Game “Find Me Save Me” to Preserve Plant Diversity in Indonesia

The current teaching materials are still in the form of books, it makes less attractive to readers. This research makes teaching materials packaged in games that invite prospective teachers to carry out Experiencing, Interaction,communication and Reflection activities in plant taxonomy learning. The purpose of this research is to develop MIKiR teaching materials from the “Find Me Save Me” game containing Indonesian plan biodiversity. This research is a development research which is analyzed descriptively. Purposive Sampling Technique, the subject of this research is the prospective science teacher of State Islamic University of Sultan Syarif Kasim Riau, State Islamic Institute of Kudus and Pattimura University which was held in March-May. The <em>“Find Me Save Me”</em> game was developed using the 4D stages in R and D, namely Define, Design, Develop, and disseminate. The game is installed through the application and data retrieval using google forms. This game uses 50 types of rare Indonesian plants which are used as a source of material for the game Find Me Save Me. The results of the research Game find me save me get proper validation and can be used in learning, and effective in introducing rare Indonesian plants, especially for prospective science teachers who are not only experts in science and technology but also love and want to preserve Indonesian plants that are starting to become rare.

Utilizing Bloom's taxonomy and authentic learning principles to promote preservice teachers' pedagogical content knowledge

The global reading crisis has been widely reported. One of the reasons attributed to the reading crisis is teachers' inadequate pedagogical content knowledge (PCK) regarding the teaching of reading. In South Africa, a multilingual country, the reading dilemma and the need for improved PCK have been debated extensively. This study, therefore, explores how utilizing Bloom's taxonomy and authentic learning principles within teacher education can promote the development of PCK in teaching reading. Using an interpretivist paradigm and an exploratory qualitative research design, data were collected from 20 second-year preservice teachers. The data collection consisted of a qualitative open-ended reflection questionnaire and document reviews. Inductive and thematic analysis was used for data analysis. The findings highlight the importance of authentic learning and PCK in teacher education and training. Furthermore, the findings indicate how Bloom's taxonomy can serve as a possible metacognitive framework to promote preservice teachers' PCK development. The benefits and challenges of PCK within teacher education are also elaborated on. Recommendations are made for teacher education programs to explore the value of Bloom's taxonomy as a metacognitive learning framework instead of a hierarchical assessment framework to create authentic learning opportunities to better prepare preservice teachers for practice.

Machine Learning Algorithm for Malware Detection: Taxonomy, Current Challenges, and Future Directions

Malware has emerged as a cyber security threat that continuously changes to target computer systems, smart devices, and extensive networks with the development of information technologies. As a result, malware detection has always been a major worry and a difficult issue, owing to shortcomings in performance accuracy, analysis type, and malware detection approaches that fail to identify unexpected malware attacks. This paper seeks to conduct a thorough literature review and offer a taxonomy of machine learning methods for malware detection that considers these problems by analyzing 77 chosen research works. The research investigates malware and machine learning in the context of cybersecurity, including malware detection taxonomy and machine learning algorithm classification into numerous categories. Furthermore, the taxonomy was used to evaluate the most recent machine learning algorithm and analysis. The paper also examines the obstacles and associated concerns encountered in malware detection and potential remedies. Finally, to address the related issues that would motivate researchers in their future work, an empirical study was utilized to assess the performance of several machine learning algorithms.

Music mood taxonomy generation and classification of Christian Kokborok song: an audio-based approach

Music mood taxonomy generation and classification of Christian Kokborok song: an audio-based approach

Interpretable Taxonomy Extraction from Digital Assets Metadata Using Automated Unsupervised Decision Tree Learning

Interpretable Taxonomy Extraction from Digital Assets Metadata Using Automated Unsupervised Decision Tree Learning

Educational Applications of Non-Fungible Token (NFT)

With the emergence of non-fungible tokens (NFTs) in blockchain technology, educational institutions have been able to use NFTs to reward students. This is done by automatically processing transaction information and the buying and selling process using smart contract technology. The technology enables the establishment of recognition levels and incentivizes students to receive NFT recognition rewards. According to the Taxonomy Learning Pyramid, learning through hands-on experiences plays a crucial role in attracting students’ interest. In this study, we analyzed the potential for using NFTs in education and the current applications of NFTs in society. We conducted a case study and performed a preliminary investigation of the types of NFT applications in the education industry. We then analyzed different education industries using individual analysis combined with SWOT analysis to understand the impact, value, and challenges of NFT applications. The results revealed 10 educational applications of NFT: textbooks; micro-certificates; transcripts and records; scholarships and rights; master classes and content creation; learning experiences; registration and data collection; patents, innovation, and research; art; payment; and deposit. Finally, ways to reduce the negative impact of education NFTs on the sustainable environment are discussed.