Automatic Concept Extraction Research Articles

ECLAD: Extracting Concepts with Local Aggregated Descriptors

Convolutional neural networks (CNNs) are increasingly being used in critical systems, where robustness and alignment are crucial. In this context, the field of explainable artificial intelligence has proposed the generation of high-level explanations of the prediction process of CNNs through concept extraction. While these methods can detect whether or not a concept is present in an image, they are unable to determine its location. What is more, a fair comparison of such approaches is difficult due to a lack of proper validation procedures. To address these issues, we propose a novel method for automatic concept extraction and localization based on representations obtained through pixel-wise aggregations of CNN activation maps. Further, we introduce a process for the quantitative comparison and validation of concept-extraction techniques based on synthetic datasets with pixel-wise annotations of their main components, mitigating possible confirmation biases induced by human visual inspection. Extensive experimentation on both synthetic and real-world datasets demonstrates that our method outperforms state-of-the-art alternatives.

Scale-preserving automatic concept extraction (SPACE)

Convolutional Neural Networks (CNN) have become a common choice for industrial quality control, as well as other critical applications in the Industry 4.0. When these CNNs behave in ways unexpected to human users or developers, severe consequences can arise, such as economic losses or an increased risk to human life. Concept extraction techniques can be applied to increase the reliability and transparency of CNNs through generating global explanations for trained neural network models. The decisive features of image datasets in quality control often depend on the feature’s scale; for example, the size of a hole or an edge. However, existing concept extraction methods do not correctly represent scale, which leads to problems interpreting these models as we show herein. To address this issue, we introduce the Scale-Preserving Automatic Concept Extraction (SPACE) algorithm, as a state-of-the-art alternative concept extraction technique for CNNs, focused on industrial applications. SPACE is specifically designed to overcome the aforementioned problems by avoiding scale changes throughout the concept extraction process. SPACE proposes an approach based on square slices of input images, which are selected and then tiled before being clustered into concepts. Our method provides explanations of the models’ decision-making process in the form of human-understandable concepts. We evaluate SPACE on three image classification datasets in the context of industrial quality control. Through experimental results, we illustrate how SPACE outperforms other methods and provides actionable insights on the decision mechanisms of CNNs. Finally, code for the implementation of SPACE is provided.

CRAFT: Concept Recursive Activation FacTorization for Explainability.

Attribution methods, which employ heatmaps to identify the most influential regions of an image that impact model decisions, have gained widespread popularity as a type of explainability method. However, recent research has exposed the limited practical value of these methods, attributed in part to their narrow focus on the most prominent regions of an image - revealing "where" the model looks, but failing to elucidate "what" the model sees in those areas. In this work, we try to fill in this gap with CRAFT - a novel approach to identify both "what" and "where" by generating concept-based explanations. We introduce 3 new ingredients to the automatic concept extraction literature: (i) a recursive strategy to detect and decompose concepts across layers, (ii) a novel method for a more faithful estimation of concept importance using Sobol indices, and (iii) the use of implicit differentiation to unlock Concept Attribution Maps. We conduct both human and computer vision experiments to demonstrate the benefits of the proposed approach. We show that the proposed concept importance estimation technique is more faithful to the model than previous methods. When evaluating the usefulness of the method for human experimenters on a human-centered utility benchmark, we find that our approach significantly improves on two of the three test scenarios. Our code is freely available: github.com/deel-ai/Craft.

Text-mined Data Improves Machine Learning Predictions for Detecting Inborn Errors of Immunity

Patients with Inborn errors of immunity (IEI) often suffer protracted diagnostic odysseys; however, machine learning (ML) can facilitate patient detection via analysis of electronic health record (EHR) data. Despite availability of structured EHR data for analytic approaches, clinical-note text represents an untapped opportunity for expanding the computable phenotype of IEI to facilitate patient-finding. This study aims to quantify text-mined feature utility for improving ML-guided IEI prediction. We mined relevant clinical concepts using EHR text from over 6,000 verified IEI patients and over 28,000 controls. Enriched IEI concepts were used to augment a Bayesian Network (probabilistic ML-model). The ML-model was then tested by analyzing EHR clinical note text from 10 verified IEI patients. From this IEI test cohort, we extracted clinical note terms in two ways - manually and via an automated Human Phenotype Ontology (HPO) extraction tool. Terms were then mapped to IEI-relevant concepts for each subject and fed into the ML-model. Prediction improvement was assessed by comparing algorithmic sensitivity for both concept extraction approaches and for the degree to which text-derived concepts enhanced predictions over structured data alone.Text-mining identified 11 additional IEI-relevant concepts. Addition of these concepts improved algorithmic sensitivity from 0.8 to 1 and enabled detection of 2 patients not previously identified. Mean cohort risk score difference (baseline vs text-enhanced ML models) was significant for both manual and automatic concept extraction experiments (0.46 ± 0.40 vs. 0.94 ± 0.10 p = 0.004 and 0.51 ± 0.44 vs. 0.85 ± 0.28 p = 0.05; threshold = 0.06). The concordance rate between expert-determined text concepts and automatically extracted HPO terms was 21% (11/47 concepts) and concept frequency distribution differences between 'expert-derived’ and 'automated-HPO’ was significant for 'Exam Findings '(p = 0.021) and 'Failure to Thrive '(p = 0.011). [Display omitted] Top concepts extracted from text include exam findings, fatigue, failure to thrive, laboratory results, medications and infections which improved ML-model diagnostic sensitivity by 20%. Here, we show that structured and unstructured EHR information allowed for the prediction of IEI in all test subjects. In addition, automated extraction of HPO terms synergistically enhances manually extracted IEI features. In summary, text-mining facilitates discovery of IEI relevant concepts and ML-driven risk prediction.

A hybrid deep learning approach for phenotype prediction from clinical notes

Identifying patient cohorts from clinical notes in secondary electronic health records is a fundamental task in clinical information management. However, with the growing number of clinical notes, it becomes challenging to analyze the data manually for phenotype detection. Automatic extraction of clinical concepts would helps to identify the patient phenotypes correctly. This paper proposes a novel hybrid model for automatically extracting patient phenotypes using natural language processing and deep learning models to determine the patient phenotypes without dictionaries and human intervention. The model is based on a neural bidirectional sequence model (BiLSTM or BiGRU) and a CNN layer for phenotypes identification. An extra CNN layer is run parallel to the hybrid model to extract more features related to each phenotype. We used pre-trained embeddings such as FastText and Word2vec separately as the input layers to evaluate other embedding's performance. Experimental results using MIMIC III database in internal comparison demonstrate that the proposed model achieved significant performance improvement over existing models. The enhanced version of our model with an extra CNN layer obtained a relatively higher F1-score than the original hybrid model. We also showed that BiGRU layer with FastText embedding had better performance than BiLSTM layer to identify patient phenotypes.

Clinical Concept Extraction with Lexical Semantics to Support Automatic Annotation.

Extracting clinical concepts, such as problems, diagnosis, and treatment, from unstructured clinical narrative documents enables data-driven approaches such as machine and deep learning to support advanced applications such as clinical decision-support systems, the assessment of disease progression, and the intelligent analysis of treatment efficacy. Various tools such as cTAKES, Sophia, MetaMap, and other rules-based approaches and algorithms have been used for automatic concept extraction. Recently, machine- and deep-learning approaches have been used to extract, classify, and accurately annotate terms and phrases. However, the requirement of an annotated dataset, which is labor-intensive, impedes the success of data-driven approaches. A rule-based mechanism could support the process of annotation, but existing rule-based approaches fail to adequately capture contextual, syntactic, and semantic patterns. This study intends to introduce a comprehensive rule-based system that automatically extracts clinical concepts from unstructured narratives with higher accuracy and transparency. The proposed system is a pipelined approach, capable of recognizing clinical concepts of three types, problem, treatment, and test, in the dataset collected from a published repository as a part of the I2b2 challenge 2010. The system’s performance is compared with that of three existing systems: Quick UMLS, BIO-CRF, and the Rules (i2b2) model. Compared to the baseline systems, the average F1-score of 72.94% was found to be 13% better than Quick UMLS, 3% better than BIO CRF, and 30.1% better than the Rules (i2b2) model. Individually, the system performance was noticeably higher for problem-related concepts, with an F1-score of 80.45%, followed by treatment-related concepts and test-related concepts, with F1-scores of 76.06% and 55.3%, respectively. The proposed methodology significantly improves the performance of concept extraction from unstructured clinical narratives by exploiting the linguistic and lexical semantic features. The approach can ease the automatic annotation process of clinical data, which ultimately improves the performance of supervised data-driven applications trained with these data.

Semi‐supervised, knowledge‐integrated pattern learning approach for fact extraction from judicial text

AbstractTremendous growth in the availability of judicial documents has demanded the rise of information extraction (IE) techniques that support the automatic extraction of relevant concepts or data from judicial texts. Among various approaches available for IE, ontology‐based IE has proven to be the most appropriate for extracting domain‐specific information from natural language text. Through this article, we propose a knowledge‐driven, semi‐supervised pattern‐based learning (bootstrapping) approach to extract domain‐specific facts from judicial text, starting with a small set of seed facts. In the semantic analysis of legal text, fact extraction is the next step to entity identification, which involves the identification of roles played by each entity in the judicial text. The proposed methodology learns extraction patterns for 12 classes of facts from the judicial text through the integration of the domain ontology called judicial case ontology (JCO). The experimental results were evaluated by human judges and found to be quite promising. One main feature of the proposed methodology is its portability across various domains (such as medical, banking, insurance, etc.) which in turn helps build expert systems in various sectors.

Automatic Concept Extraction for Domain and Student Modeling in Adaptive Textbooks

The increasing popularity of digital textbooks as a new learning media has resulted in a growing interest in developing a new generation of adaptive textbooks that can help readers to learn better through adapting to the readers’ learning goals and the current state of knowledge. These adaptive textbooks are most frequently powered by internal knowledge models, which associate a list of unique domain knowledge concepts with each section of the textbook. With this kind of concept-level knowledge representation, a number of intelligent operations could be performed, which include student modeling, adaptive navigation support, and content recommendation. However, manual indexing of each textbook section with concepts is challenging, time-consuming, and prone to errors. Modern research in the area of natural language processing offers an attractive alternative, called automatic keyphrase extraction. While a range of keyphrase and concept extraction methods have been developed over the last twenty years, few of the known approaches were applied and evaluated in a textbook context. In this paper, we present FACE, a supervised feature-based machine learning method for automatic concept extractions from digital textbooks. This method has been created for building domain and student models that form the core of intelligent textbooks. We evaluated FACE on a newly constructed full-scale dataset by assessing how well it approximates concept annotations produced by human experts and how well it supports the needs of student modeling. The results show that FACE outperforms several state-of-the-art keyphrase extraction methods.

Online education empowerment with artificial intelligence tools

The experience of organizing the educational process during the quarantine caused by the COVID-19 pandemic is considered. Using of interactive technologies that allow organizing instant audio communication with a remote audience, as well as intelligent tools based on artificial intelligence that can help educational institutions to work more efficiently. Examples of sufficient use of artificial intelligence in distance learning are given. Particular attention is paid to the development of intelligent chatbots intended for use in communications with students of online courses of educational web portals. The use of technologies of ontology formation based on automatic extraction of concepts from external sources is offered, what can lead to greater acceleration of construction of the intellectual component of chatbots. Artificial intelligence tools can become an essential part of distance learning during this global COVID-19 pandemic. While educational institutions are closed to quarantine and many of them transitioned to distance learning lecturers and schoolteachers, as well as students and schoolchildren faced with the necessity to study in this new reality. The impact of these changes depends on people's ability to learn and on the role that the education system will play in meeting the demand for quality and affordable training. The experience of organizing the educational process at the University of Education Management of the National Academy of Pedagogical Sciences of Ukraine in the quarantine caused by the COVID-19 pandemic showed that higher and postgraduate institutions were mostly ready to move to distance learning. However, most distance learning systems, on whatever platform they are organized, need to be supplemented: the ability to broadcast video (at least ‒ one-way streaming), providing fast transmission of various types of information, receiving instant feedback when voting, polls and more. The structure of each section of the training course for the online learning system should fully cover the training material and meet all the objectives of the course. Appropriate language should be used, and wording, syntax, and presentation of tasks should be considered. One of the areas of application of artificial intelligence technologies in online learning is the use of chatbots which are characterrized by the following properties. It is advisable to use computer ontologies to ensure the intellectualization of chatbots. In this case, the metadata must be understandable to both humans and software and meet the requirements of modern standards in the field of information technology. The extraction of concepts from external data sources was carried out to build the ontology.

Automatic extraction of Uyghur domain concepts based on multi‐feature for ontology extension

In the internet age, as a conceptual model of knowledge organisation, ontology has become a research hotspot. Ontology extension achieves the purpose of expanding ontology by adding new concepts and discovering the relationships between concepts in the existing ontology. In order to improve the automation and accuracy of ontology concept extraction in the Uyghur language, here the authors propose a new method to automatically extract concepts from text collection. For the characteristics of the Uyghur domain ontology concept, the text preprocessing is performed first, then the inter-word correlation of multi-feature fusion is calculated, such as Mi, Cd and Ea, and finally, the domain terminology and concept are automatically extracted, according to the term frequency–inverse document frequency algorithm. Experiment results show that, in terms of precision and recall rate, the multi-feature method proposed here represents an improvement over other methods. It also proves the feasibility and effectiveness of the authors’ method to extract the domain concept.

Automatic Concept Extraction Based on Semantic Graphs From Big Data in Smart City

With the rapid development of smart cities, various types of sensors can rapidly collect a large amount of data, and it becomes increasingly important to discover effective knowledge and process information from massive amounts of data. Currently, in the field of knowledge engineering, knowledge graphs, especially domain knowledge graphs, play important roles and become the infrastructure of Internet knowledge-driven intelligent applications. Domain concept extraction is critical to the construction of domain knowledge graphs. Although there have been some works that have extracted concepts, semantic information has not been fully used. However, the excellent concept extraction results can be obtained by making full use of semantic information. In this article, a novel concept extraction method, Semantic Graph-Based Concept Extraction (SGCCE), is proposed. First, the similarities between terms are calculated using the word co-occurrence, the LDA topic model and Word2Vec. Then, a semantic graph of terms is constructed based on the similarities between the terms. Finally, according to the semantic graph of the terms, community detection algorithms are used to divide the terms into different communities where each community acts as a concept. In the experiments, we compare the concept extraction results that are obtained by different community detection algorithms to analyze the different semantic graphs. The experimental results show the effectiveness of our proposed method. This method can effectively use semantic information, and the results of the concept extraction are better from domain big data in smart cities.

Aplicação de revisão sistemática com suporte de mineração de dados e de textos: o caso do periódico Design Studies

O emprego de revisão sistemática na academia tem apresentado crescimento considerável. Frequentemente, esta tem sido usada como uma forma de filtragem por meio de palavras-chave. Contudo, é levantada, nesse processo, uma grande quantidade de dados, a partir da qual se acredita que possam ser identificadas relações mais aprofundadas, empregando técnicas de mineração de dados ou textos. Este estudo tem como objetivo identificar e verificar relações entre tópicos de pesquisa, considerando como uma prova de conceito os artigos do periódico Design Studies, por meio da aplicação de tecnologias de Descoberta de Conhecimentos em Bases de Dados. Em uma abordagem quantitativa de alcance descritivo, apresenta-se, inicialmente, uma caracterização da amostra de 483 artigos desse periódico a partir de estatística descritiva, abarcando as palavras-chave provenientes dos seus metadados. Em seguida, é usada a ferramenta de mineração de texto AlchemyAPI, para extração automática de conceitos e palavras-chave, e a ferramenta RapidMiner Studio, para análise de associação a partir dos metadados e dos resultados da mineração de texto. Constatou-se a recorrência de tópicos como o Processo, a Cognição, a Criatividade e a Educação em Design. Foi possível verificar, por meio da ferramenta AlchemyAPI, conceitos que não foram expressos nas palavras-chave dos artigos, o que pode ser considerado como uma potencialidade para estudos de revisão sistemática. Assim, espera-se que as técnicas aqui abordadas possam ser mais exploradas em pesquisas de diversas áreas, enquanto suporte válido para revisões sistemáticas de publicações científicas.

A Lexicon based Approach Towards Concept Extraction

The emergence of digital media has tremendously increased amount of unstructured data. Recently 80% of data, generated over web, is in unstructured format. This immense amount of data is a great source for the knowledge discovery and thus, may be utilized for extracting purposeful information. This study adopted lexicon based approach for automatic concept extraction from online news stories and events. An application prototype has been developed to demonstrate the applicability and effectiveness of the adopted approach. The extracted knowledge about news stories, articles and blogs is essential in understanding in-depth information for news analysts. This knowledge plays vital role towards building societies, since media is considered as an opinion maker for its audience.

Automated two-phase business model-driven synthesis of conceptual database models

Existing approaches to business process model-driven synthesis of data models are characterized by a direct synthesis of a target model based on source models represented by concrete notations, where the synthesis is supported by monolithic (semi)automatic transformation programs. This article presents an approach to automated two-phase business process model-driven synthesis of conceptual database models. It is based on the introduction of a domain specific language (DSL) as an intermediate layer between different source notations and the target notation, which splits the synthesis into two phases: (i) automatic extraction of specific concepts from the source model and their DSL-based representation, and (ii) automated generation of the target model based on the DSL-based representation of the extracted concepts. The proposed approach enables development of modular transformation tools for automatic synthesis of the target model based on business process models represented by different concrete notations. In this article we present an online generator, which implements the proposed approach. The generator is implemented as a web-based, service-oriented tool, which enables automatic generation of the initial conceptual database model represented by the UML class diagram, based on business models represented by two concrete notations.

Key Concept Identification: A Comprehensive Analysis of Frequency and Topical Graph-Based Approaches

Automatic key concept extraction from text is the main challenging task in information extraction, information retrieval and digital libraries, ontology learning, and text analysis. The statistical frequency and topical graph-based ranking are the two kinds of potentially powerful and leading unsupervised approaches in this area, devised to address the problem. To utilize the potential of these approaches and improve key concept identification, a comprehensive performance analysis of these approaches on datasets from different domains is needed. The objective of the study presented in this paper is to perform a comprehensive empirical analysis of selected frequency and topical graph-based algorithms for key concept extraction on three different datasets, to identify the major sources of error in these approaches. For experimental analysis, we have selected TF-IDF, KP-Miner and TopicRank. Three major sources of error, i.e., frequency errors, syntactical errors and semantical errors, and the factors that contribute to these errors are identified. Analysis of the results reveals that performance of the selected approaches is significantly degraded by these errors. These findings can help us develop an intelligent solution for key concept extraction in the future.

Automatic Extraction of Learning Concepts from Exam Query Repositories

Educational data mining (or EDM) is an emerging interdisciplinary research field concerned with developing methods for exploring the specific and diverse data encountered in the field of education. One of the most valuable data sources in the educational domain are exam query repositories, which are commonly pre-dating modern e-learning systems. Exam queries in those repositories usually lack additional metadata which helps establish relationships between questions and corresponding learning concepts whose adoption is being tested. In this paper we present our novel approach of using data mining methods able to automatically annotate pre-existing exam queries with information about learning concepts they relate to, leveraging both textual and visual information contained in the queries. This enables automatic categorization of exam queries which allows for both better insight into the usability of the current exam query corpus as well as easier reporting of learning concept adoption after these queries are used in exams. We apply this approach to real-life exam questions from a high education university course and show validation of our results performed in consultation with experts from the educational domain.

Abstract 5283: Shangri-Docs: a browser based tool for document exploration and automatic knowledge extraction from unstructured biomedical text

Abstract Biomedical information is available to research and development scientists as unstructured text in the form of scientific manuscripts and reports published in the literature and elsewhere. Scientists focused on specific research programs are burdened with surveying vast numbers of publications and reports to acquire information relevant to their efforts. Employing technology as a research aid provides a mechanism to cope with information overload that characterizes the R&D environment. Text mining can extract knowledge from large corpora of biomedical text and make it available to support scientific research and knowledge collections [1, 2] and intelligent PDF reader tools able to search content and find related articles [3] are available; however, such reader tools are typically desktop applications limited to specific platforms and data sources so they cannot easily support broad based integrated scientific search needs for a dispersed R&D organization with a wide variety of content needs. Our team has developed a web-browser based document reader with a built-in exploration tool and automatic concept extraction from biomedical text content. This provides R&D scientists with a simple tool to aid finding, reading, and exploring documents relevant to focused research objectives. The tool, Shangri-Docs, combines a document reader with automatic concept extraction and highlighting of relevant terms based on carefully selected ontologies combined with our custom corporate enterprise taxonomy. Shangri-Docs provides the ability to evaluate a wide variety of document formats (e.g. PDF, Word, PPT, text, etc.) and exploits the linked nature of the Web and personal content by performing searches on content from public sites (e.g. Wikipedia, PubMed) and privately cataloged databases simultaneously. Shangri-Docs incorporates Apache cTAKES (clinical Text Analysis and Knowledge Extraction System) [4] and Unified Medical Language System (UMLS) to automatically identify and highlight terms and concepts, such as specific pathology, disease, drug, and biological terms mentioned in the text. cTAKES was originally designed specifically to extract information from clinical medical records. We have extended cTAKES automatic knowledge extraction process to include the R&D biomedical research domain by improving the ontology guided information extraction process. Shangri-Docs could be adapted to other science fields and further customized across our R&D scientific community via our open source, cloud-based, data management system. [1] Funk, et.al., BMC Bioinformatics 2014, 15:59 doi:10.1186/1471-2105-15-59 [2] Kang et al., BMC Bioinformatics 2014, 15:64 doi:10.1186/1471-2105-15-64 [3] Utopia Documents, http://utopiadocs.com [4] Apache cTAKES, http://ctakes.apache.org Citation Format: Chris Mattmann, Lauren Intagliata, Selina Chu, Garth McGrath, Giuseppe Totaro, Daniel Civello, David Ballard, Jeffrey Long, Nipurn Doshi, Shivika Thapar, Michael Livstone, Paul Ramirez, Maureen Cronin. Shangri-Docs: a browser based tool for document exploration and automatic knowledge extraction from unstructured biomedical text. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 5283.

Automatic detection of expert models: The exploration of expert modeling methods applicable to technology-based assessment and instruction

This mixed methods study explores automatic methods for expert model construction using multiple textual explanations of a problem situation. In particular, this study focuses on the key concepts of an expert model. While an expert understanding of a complex problem situation provides critical reference points for evidence-based formative assessment and feedback, the extraction of those reference points has proven challenging. Building upon semantic analysis, this study utilizes deep natural language processing techniques to facilitate the automatic extraction of key concepts from textual explanations written by experts. The study addresses the following question: (a) whether experts in a domain represent a common understanding of a problem situation through shared key concepts, (b) which metrics extract key concepts from textual data most accurately, and (c) whether automatic methods enable expert model construction from a corpus of textual explanations instead of a pre-defined, ideal explanation created using the Delphi method. The OntoCmap tool was used to extract concepts from multiple textual explanations and to generate diverse metrics assigned to each concept. The findings indicate that (a) experts have varying ways of understanding a problem situation, (b) graph-based filtering metrics (i.e., betweenness and reachability) performed better in building a set of key concepts, and (c) a single, pre-defined explanation led to a more accurate set of key concepts than a corpus of explanations from various experts.

Event Detection and Summarization in Soccer Videos Using Bayesian Network and Copula

Semantic video analysis and automatic concept extraction play an important role in several applications; including content-based search engines, video indexing, and video summarization. As the Bayesian network is a powerful tool for learning complex patterns, a novel Bayesian network-based method is proposed for automatic event detection and summarization in soccer videos. The proposed method includes efficient algorithms for shot boundary detection, shot view classification, mid-level visual feature extraction, and construction of the related Bayesian network. The method contains of three main stages. In the first stage, the shot boundaries are detected. Using the hidden Markov model, the video is segmented into large and meaningful semantic units, called play-break sequences. In the next stage, several features are extracted from each of these units. Finally, in the last stage, in order to achieve high level semantic features (events and concepts), the Bayesian network is used. The basic part of the method is constructing the Bayesian network, for which the structure is estimated using the Chow-Liu tree. The joint distributions of random variables of the network are modeled by applying the Farlie-Gumbel-Morgenstern family of Copulas. The performance of the proposed method is evaluated on a dataset with about 9 h of soccer videos. The method is capable of detecting seven different events in soccer videos; namely, goal, card, goal attempt, corner, foul, offside, and nonhighlights. Experimental results show the effectiveness and robustness of the proposed method on detecting these events.

Test Cases Selection Based on Source Code Features Extraction

Extracting valuable information from source code automatically was the subject of many research papers. Such information can be used for document traceability, concept or feature extraction, etc. In this paper, we used an Information Retrieval (IR) technique: Latent Semantic Indexing (LSI) for the automatic extraction of source code concepts for the purpose of test cases’ reduction. We used and updated the open source FLAT Eclipse add on to try several code stemming approaches. The goal is to check the best approach to extract code concepts that can improve the process of test cases’ selection or reduction.