Search

Prediction is widely used for various purposes and in many fields of human activity. The techniques employed for making predictions are a subject of great scientific interest within the research community due to their diversity, level of accuracy, and adaptability to data. The challenge is to determine the factors that affect the choice of an optimal technique suited to each prediction objective. In this article, we conduct a review of models used in the literature to make predictions in different domains to understand the factors influencing the selection of a specific predictive model in relation to their areas of study. A comparative analysis of prediction techniques such as statistical algorithms, Data Mining, and Machine Learning has been performed. It follows that the selection of an adequate prediction technique for the best decision-making should take into account the projection horizon, uncertainty around the prediction, data availability and reliability, and the associated cost of prediction.

Understanding consumers’ preferences has always been important for economic theory and for business practitioners in operations management, supply chain management, marketing, etc. While preferences are often considered stable in simplified theoretical modelling, this is not the case in real-world decision-making. Therefore, it is crucial to understand consumers’ preferences when a market disruption occurs. This research aims to recognise consumers’ preferences with respect to online shopping after the COVID-19 outbreak hit markets. To this purpose, we conducted an empirical study among Croatian consumers with prior experience in online shopping using an online questionnaire. The questionnaire was completed by 350 respondents who met the criteria. We selected decision-tree models using the J48 algorithm to determine the influences of the found shopping factors and demographic characteristics on a consumer’s preference indicator. The main components of our indicators that influence consumer behaviour are the stimulators and destimulators of online shopping and the importance of social incidence. Our results show significant differences between men and women, with men tending to use fewer variables to make decisions. In addition, the analysis revealed that four product groups and a range of shopping mode-specific influencing factors are required to evaluate consumers’ purchase points when constructing the consumers’ preference indicator.

The article reflects the current trends in working with the digital heritage of Russian literature, examines the process of forming virtual archives as a gradual accumulation of the “big data” of scientific research, i. e. unrecognized information array of raster documents containing tens of thousands of images. The research analyzes the specifics of scientific work in the field of ego-documentary heritage that arose at the turn of the 20th – 21st centuries (a corpus of diary entries, workbooks, notebooks, correspondence), the principles of publication and modern standards of digitization of archival heritage. The study and practicing of the three most promising virtual resources on the history of Russian literature of the mid-19th – first half of the 20th centuries allows to formulate specific tasks and methods of visualization of a large corpus of raster images of archival documents, as well as previously untapped possibilities of search engine automation. Much attention is paid to the transition from the graphical elements of the raster image of the manuscript to semantic ones, which allow the use of data mining elements for an unrecognized data array.

Cette étude propose à l’IMF ALPHA, ainsi surnommée pour des raisons d’anonymat et de respect de sa politique de confidentialité, une approche et un outil efficace pour fiabiliser davantage son processus d’octroi des prêts, jugé insuffisamment efficace.

The digital transformation of construction and housing and utilities sector involves data-based management. The most accessible data is telemetry of electricity, gas, heat and water consumption. Despite the fact that not all apartment buildings are equipped with meters yet, it is necessary to think about methods for processing measurement data. Intelligent data processing methods are gaining wide popularity. The purpose of this study is the processing and analysis of water consumption telemetry data. The objective of the study is to determine the periods of nighttime water consumption using the clustering method. The data of apartment meters of hourly consumption of hot and cold water are investigated. The measurement period is 1 month. Intelligent cluster analysis was conducted based on the DBSCAN machine learning model (Density-based spatial clustering of applications with noise). Clustering objects are the hours of the day. As a result of the study, the hours of night consumption of cold and hot water were allocated, both in the whole month and separately on weekdays and Sundays. The conclusion is made about the benefits of using intelligent cluster analysis of water consumption telemetry data for effective management of water resources and equipment.

This study aims to predict the sales of a commercial organization in order to know the role that modern information technology plays in achieving accurate and rapid processing of data based on the data mining tool represented in the Microsoft Power BI business intelligence program, through a theoretical and applied study. The significant role played by the estimated future sales information in the planning process as well as guiding and rationalizing the decisions of the sales manager to improve the performance of the organization.

Data mining and AI techniques are increasingly being used to automate data analysis. Ideally, one may wish to completely automate the data analysis process, but in many real-world applications a full automation may pose significant risks. In these cases, human analysts must be directly involved to refine the analysis or to make the final decisions. A challenging problem, therefore, is how to perform efficient and trustworthy decision-making when humans are an integral part of the analysis pipeline. We propose a “human-in-the-loop” methodology that leverages data mining, machine learning, and visual analytics to improve and speed up the analysis. A key feature is the use of a dashboard that integrates intuitive visual tools, which aid analysts to efficiently discover hidden data patterns or to get helpful insights. We describe in particular how this methodology has been successfully applied to support Revenue Agency officers in tax risk assessment.

Data Mining es una tecnología de estadística computarizada para la extracción de pautas reiteradas y significativas a partir de datos cuantiosos. En este artículo se exponen primero los fundamentos para su aplicación al análisis de los textos. Después, se detalla la aplicación del análisis de afinidad a buena parte de los escritos de Tomás de Aquino. Y, finalmente, se muestran los resultados obtenidos del análisis de varios de sus comentarios paulinos.

Continous development of web, social networking, mobile computing and Internet of Things in the last decades has led to a dramatic increase in the volume of data generated and, consequently, to the need for new ways of storing and analyzing of these data. In addition, these data no longer have a structured format but are rather in a semistructured or unstructured format. They contain useful information that companies need to explore. So, the methods of mining structured data had to be developed to mining the semistructured and also the unstructured data. In this paper we describe some of the most fundamental mining tasks.

By studying the figures and media that inform the machine imagined by Alex Garland in DEVS, this article proposes an archaeology of this invention that lies at the crossroads of various problems of contemporary media theory and that aims to answer, in the words of speculative science fiction, the question of spatiotemporal ubiquity, inspired by a precise technological context: Californian tech.

ABSTRACT Governments face a tough and timeless challenge: dealing with the capability of radical terrorist organizations to recruit foreign fighters. However, scholars so far have ignored that this phenomenon pertains to the realm of complexity theory, failing to determine the combination of country-level variables able to catalyze this issue. This is an important concern if countries want to design effective socio-political strategies aimed at decreasing terrorist groups’ capability to enroll foreign fighters or, at least, to curtail the penetration of their radical message. Thus, to address this issue we undertake an exploratory data mining approach (knowledge discovery in databases) to discover country-level patterns which might engender conditions that induce people to join an extremist organization, based on the case of ISIS. After a pre-selection procedure, the 950 variables initially selected were reduced to 22, and subsequently used in decision tree algorithms. Findings reveal the existence of six specific country clusters, which are characterized by some spatial, structural (economic and political), and social variables that create favorable conditions for the emergence of the phenomenon. Academic and practical recommendations are then discussed.

The digitalization of the economy is a trendy phenomenon that is transforming the micro and macro-economy, evolving at a record pace and directly impacting the performance and social entrepreneurship of organizations in various sectors around the world. Organizations must therefore face vast waves of data to follow the news and facilitate territorial intelligence. This world of data is defined by the collective name of data mining, the processing of which is usually carried out by artificial intelligence technology. Entrepreneurship creates spaces of solidarity, mutual aid and coalition between social groups suffering from poverty, precariousness and even precariousness, with the aim of ensuring equitable distribution for the benefit of all. In this context, Morocco has embarked on structural changes, strengthened its modern and competitive economy, facilitated the creation of growth and encouraged entrepreneurship, and has committed itself to national commitments to territorial development. To get a broader and clearer idea of the impact of data mining and economic intelligence on territorial intelligence and social entrepreneurship, we seek to answer the following questions: How does social entrepreneurship promote territorial dynamics through artificial intelligence, particularly in the Fez-Meknes region?

In the literature, new machine learning algorithms are dynamically produced in the field of artificial intelligence engineering and the algorithms are constantly updated with new parameter estimations. The performance of existing algorithms in various business areas is still an important topic of discussion. Also, machine learning algorithms are frequently used in long-term credit ratings, which is an crucially important sub-branch of finance. This study was conducted to determine which popular machine learning model performs better in credit scoring. Artificial Neural Network, Random Forest, Support Vector Machine and K Nearest Neighbor were used to determine the algorithm that is suitable for the structure, attribute content and distribution of the data, and the operating logic of the models. In the study, the long-term credit rating is the target variable and the remaining variables are the features, the prediction performances of these 4 algorithm, which are frequently used in previous studies such as credit rating, credit risk, fraud analysis were compared. After data preprocessing, a classification study was carried out using the features included in the model. The metrics used in the comparison are MSE, RMSE, MAE and accuracy. According to the metrics, RF algorithm showed the best performance in the credit scoring.

Credit card fraud is a major problem that has caused several challenges for practitioners in the accounting and finance industry due to a large number of daily transactions as well as the difficulties encountered in identifying fraudulent transactions. The purpose of this study is to investigate the application of artificial intelligence techniques as a fraud detection mechanism that can effectively and efficiently detect credit card fraud and identify fraudulent financial transactions. The data was acquired from 100 respondents across the accounting and finance industry and analysed using SPSS. Researcher analysed the data using regression analysis, Pearson correlation coefficient, and reliability analysis. Findings revealed that the three artificial intelligence techniques machine learning, data mining, and fuzzy logic have a significant positive relationship with credit card fraud detection. However, fuzzy logic was discovered to be the least utilized by experts due to its low accuracy/precision in comparison with machine learning and data mining. Based on these findings, our study concludes that the application of artificial intelligence techniques provides experts with better accuracy and efficiency in detecting fraudulent transactions. Therefore, it is recommended that fraud examiners, auditors, accountants, bankers, and organizations should implement and apply artificial intelligence techniques in order to spot anomalies faster and identify fraudulent financial transactions effectively and efficiently.

In times of constant and accelerated changes, the current business environment is characterized by the need to introduce knowledge in the activities developed. Organizations recognize that knowledge factors have a significant impact or added value on business performance. It is because of this need that a study on the production of spirits (aguardiente) under the case study methodology is presented in this article, by presenting a conceptual framework, research dynamics and literature review. This allows providing a theory as a proposed solution to the use, transfer and conservation of knowledge, recognition of information flows, storage and dissemination policies for the production of spirits. The proposal allows validating the theory of infoknowledge and incorporating to it a new technological tool that guarantees the integral treatment of the functionalities foreseen in the infoknowledge model, with this, the process of spirits ensures advance processing, data mining, content management, process monitoring, notifications and the operation of the network of experts

Resumo Neste artigo, discutimos um novo regime de poder, nomeado de Governamentalidade Algorítmica por Antoinette Rouvroy e Thomas Berns, que, crescentemente, vem operando na condução de nossas condutas. Diferentemente do poder disciplinar e da biopolítica, tal governamentalidade não tem por centro de gravidade os indivíduos ou as populações. Antes, por meio da mineração de dados e da produção de perfis, age tanto no nível infra-individual quanto supra-individual. Para problematizá-la, analisamos e colocamos em questão um dos seus modos de operação, os sistemas de recomendação e, dentre eles, focalizamos o caso da Netflix. Por fim, tensionamos os efeitos que os algoritmos que lhe constituem podem ter em nossos modos de subjetivação, já que, não raro, tendem a excluir de nossas experiências aquilo que é da ordem do imprevisto e que é capaz de fazer com que algo nos aconteça e nos transforme.

Cotton used as herba lraw material of textile industry is of strategic importance in our country and also in agriculture, industry and trade in the world. Cotton trade between countries takes place through cotton stock exchanges. The state determines the value of many agricultural products in our country. However, cotton is one of the few products whose prices are determined by the stock exchange. In this study, annual cotton prices were taken from Manisa Agricultural Market. Later, these data was first corrected, and analyzed in the weka program. Time series and artificial neural network techniques were used to estimate the value of pricesfor 2017. Based on the estimated valuesby MAE, MAPE and RMSE, it is decided which technique will yield better prediction performance. The results obtained were also compared with similar studies in the literature. All of these results show that the artificial neural network technique achieved more successful results.

Ce travail trouve son ancrage dans une enquête de terrain sur les mutations à l’œuvre dans les rédactions de la PQR, presse quotidienne régionale. Le datajournalisme avait alors émergé comme une notion à questionner et à redéfinir tant les discours produits semblaient complexes. Entre espoirs, souvent idéalisés, d’un outil clé en main pour illustrer l’actualité ultra-locale et verrous institutionnels et individuels éprouvés lors des pratiques, il semblait y avoir un espace complexe à explorer. Nous proposons ici de présenter le rôle particulier que tient la datavisualisation dans cet espace : souvent confondue avec le datajournalisme, la datavisualisation constitue fréquemment la première étape d’une familiarisation au traitement des données, facilitée par le recours à des outils clé en main qui interviennent à deux étapes du traitement journalistique : lors de l’exploration des données par le journaliste puis dans leur mise en narration pour le lecteur.

At present, the research on ecological civilization city construction in China is in the stage of blooming. The evaluation of ecological civilization, the indicator system and the implementation path are all in the preliminary stage of research, and a unified and complete understanding has not yet been formed. The purpose of this paper is to study the informatization construction of ecological civilization evaluation based on data mining technology. On the basis of a comprehensive understanding of the current research status of the development of ecological civilization, using the combination of subjective and objective screening methods, comprehensively considering the construction of ecological civilization, the evaluation index system suitable for the construction of ecological civilization is analyzed. This paper builds an ecological civilization evaluation system based on data mining technology, uses the system to evaluate the efficiency of ecological civilization construction in M province, and guides the construction of ecological civilization in China. Through the evaluation results, it enriches the understanding of ecological civilization construction, and scientifically grasps development advantages, providing the government with accurate and effective evaluation tools.

Internet is a basic need for everyone, including students. The internet can be used as a means to improve a student's learning achievement or vice versa, if the use of the internet is excessive. So that 3 categories were taken, including intensity, purpose of internet use and student GPA. This study aims to see the correlation between the intensity and purpose of internet use and student learning achievement which is implemented in a web-based system. The method used in this research is apriori method, which is a method for finding patterns or correlations between items. Data taken with a total of 100 data. The results of this study resulted in a combination of the goal item set, intensity and GPA along with the association rules. The highest item obtained was social media with a value of 64 and for the correlations of the correlation between intensity, goal and student GPA, the 3 highest association rules were obtained as follows: 'Online Game, Rare Intensity → Low GPA' with a confidence value 100, 'Very Frequent Intensity, Search Engine →GPA Medium' with a confidence value 85.71 and 'social media, Frequent Intensity→High GPA 'with a confidence value 41.38.

With the development of large-scale electrochemical energy storage power stations, lithium-ion batteries have unique advantages in terms of re-energy density, power density, and cycle life, and are applied to power system energy storage devices. However, behind the rapid development, there are many key issues unanswered, which are likely to lead to various safety accidents. Therefore, it is very important to conduct consistency analysis of lithium batteries used in large-scale power systems to prepare for system safety assessment. This paper mainly explains the reasons and manifestations of the inconsistency, and based on data mining algorithms, uses the charging voltage curve clustering analysis method based on subtractive clustering to evaluate the consistency of lithium-ion batteries.

Since the outbreak of COVID-19, all countries have taken certain measures to restrain the spread of COVID-19. It is a complex and practical issue to study the measures and effects of the epidemic prevention and control. The spread of the epidemic in different countries has certain characteristics, and the prevention and control models and effects in different countries are also different. In this paper, the key factors affecting the prevention effect of COVID-19 were analyzed by integrating multiple data and using interdisciplinary methods of mathematics, statistics and information management science on the basis of public management, and the system model of influencing factors of COVID-19 was constructed.

At present, China exists a problem that the cost of food sampling inspection is too high. This paper attempts to reduce the number of sampling inspection items in the same food category, reduce the cost of food sampling inspection, and improve the work efficiency through the association analysis of national sampling inspection data. And this paper applies Apriori algorithm to analyse the association rules, which is based on the unqualified pastry sampling inspection data in the 2019 national food sampling inspection database. Finally, we obtain 10 strong association rules through experiments. The results show that this association analysis can reduce the workload of food sampling inspection effectively.

This contribution is a report on the panel "Big Translation History" that took place at the Congress of the European Society for Translation Studies in Stellenbosch in September of 2019.

The so called cognitive ontology (conceptual knowledge base) of the Cognitive Atlas project (http://www.cognitiveatlas.org/) -that characterizes the current state of thinking in cognitive sciences- and the Google Ngram Viewer and other data mining tools for analysis were used to assess the philosophical-psychological conceptual transformation from the textual corpus of two classical works within philosophy (Critique of Pure Reason, Kant) and psychology (Principles of Psychology, by James), respectively; in order to ask ourselves: What proportion of philosophical concepts from 1700 are still scientifically valid in psychology in terms of their use and frequency? How many concepts that appear in the cognitive ontology (Cognitive Atlas) also appear in these classic works of Philosophy (Critique of Pure Reason, by Kant) and Psychology (Principles of Psychology, by James)? How many concepts used in these two classics works have already been dismissed for being pre-scientific or maintained? Seems to us that the use of Big Data and other tools of statistical analysis can solve with data and evidence long-debated issues on the historical origin of terms, the conceptual transformation and progress of the disciplines.

It is an era of big data nowadays. Processing, analyzing and working with data are new opportunities and challenges. This article mainly studies the development and application of employment information management platform of colleges and universities under the background of big data. As the number of graduates increases year by year, the employment situation is becoming more and more severe. The employment information management platform is playing an increasingly important role as an information exchange platform for colleges and universities to learn about the employment situation of students, and for employers to release recruitment information and for students’ job hunting. The employment information management platform makes full use of big data for analysis, which can well control the employment; at the same time, the school grasps the changing trend of society’s needs for talents through analysis, and the employment situation of graduates will be used as the main basis to revise talent training plans, formulate enrollment plans, and carry out feasibility study for adding new majors. The big data platform provides a basis for decision-making of colleges and universities and is of great significance to the development of school. Therefore, it is extremely urgent to build an efficient and convenient employment information management platform to meet the needs of colleges and universities, employers and students.

The amount of rainfall that occurs cannot be determined with certainty, but it can be predicted or estimated. In predicting the potential for rain, data mining techniques can be used by classifying data using the naive Bayes method. Naïve Bayes algorithm is a classification method using probability and statistical methods. The purpose of this study is how to implement the naive Bayes method to predict the potential for rain in Ternate City, and be able to calculate the accuracy of the Naive Bayes method from system created. The highest calculation results with new data with a total of 400 training data and 30 test data, obtained 30 correct data with 100% precision, 100% recall and 100% accuracy and the lowest calculation results with new data with a total of 500 training data and 50 test data, obtained 38 correct data and 12 incorrect data with a percentage of precision 61.29%, recall 100% and accuracy 76%.

Copyright of Traitement Automatique des Langues is the property of Association pour le Traitement Automatique des Langues (ATALA) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Customer behavior classification can be useful to assist companies in conducting business intelligence analysis. Data mining techniques can classify customer behavior using the K-Nearest Neighbor algorithm based on the customer's life cycle consisting of prospect, responder, active and former. Data used to classify include age, gender, number of donations, donation retention and number of user visits. The calculation results from 2,114 data in the classification of each customer’s category are namely active by 1.18%, prospect by 8.99%, responder by 4.26% and former by 85.57%. System accuracy using a range of K from K = 1 to K = 20 produces that the highest accuracy is 94.3731% at a value of K = 4. The results of the training data that produce a classification of user behavior can be used as a Business Intelligence analysis that is useful for companies in determining business strategies by knowing the target of optimal market.

Editoriale.

With the evolution of the large general retail sector in Morocco, out-of-stocks (OOS) have become one of the major problems retailers have to resolve in order to retain their existing customers and attract new ones. This is particularly because the rate of OOS could indeed be considered as a performance indicator when the products and services offered by the competing retailers are relatively the same. This paper provides a frame of reference in relation to the concept of “stock-out” in addition to the information and communication technologies used to improve some of the processes involved in the retail supply chain. We then present the results of a field study, which we have conducted in the large retail sector in Morocco in order to identify the scope of use of such technologies with the aim of reducing OOS situations at the store level.

Cet article décrit la conception et la mise à l’essai d’un programme informatique de production semi-automatisée de cartes conceptuelles. Cette technique dérivée de l’extraction d’information a pour but de produire une représentation simple et signifiante du contenu d’un ou de plusieurs textes sous forme d’un schéma de connaissances. Le programme a été mis à l’essai à partir de deux corpus de textes choisis pour couvrir deux thèmes du Programme de formation de l’école québécoise de quatrième secondaire en science et technologie. Les résultats de l’étude sont très encourageants et montrent le potentiel d’une telle approche pour faciliter la construction d’une carte conceptuelle, une tâche qui est généralement réalisée par des humains. Les résultats suscitent également quelques réflexions quant aux modalités courantes d’élaboration et d’évaluation de cartes conceptuelles.

In this paper we briefly discuss the history of corpus linguistics and its applications to the study of scientific language. We provide some exemplification for corpora exploitation tools, corpora applications to scientific language studies, data mining and natural language processing.

Water cut is an important parameter in reservoir management and surveillance. Unlike traditional approaches, including numerical simulation and analytical techniques, which were developed for predicting water production in oil wells based on some assumptions and limitations, a new data-driven approach is proposed for forecasting water cut in two different types of oil wells in this article. First, a classification approach is presented for water cut prediction in sweet oil wells with discontinuous salt production patterns. Different classification algorithms including Support Vector Machine (SVM), Classification Tree (CT), Random Forest (RF), Multi-Layer Perceptron (MLP), Linear Discriminant Analysis (LDA) and Naïve Bayes (NB) are investigated in this regard. According to the results of a case study on a real Iranian sweet oil well, RF, CT, MLP and SVM can provide the best performance measures, respectively. Next, a Vector Autoregressive (VAR) model is proposed for forecasting water cut in salty oil wells with continuous water production during the life of the well. The proposed VAR model is verified using data of two real salty oil wells. The results confirm that the well-tuned proposed VAR model could provide reliable and acceptable results with very good accuracy in forecasting water production for the near future days.

Higher education has an important role to develop human resources in the economic growth and development of the country. One of specific way of evaluating and analyze data in education is to use data mining techniques. C4.5 algorithm as one of the data mining techniques that have good performance is very relevant used for data analysis tools. In this research using data on the performance of lecturers in college, there are 100 records with a 6 variable that affects individual factors in the productivity of lecturers including age, employment, attendance, certification, position, Education, and additional duties. In the end of the mining result, the forward chaining method is used to extract the rules that are generated by C4.5 algorithm. The input premises are examined by forwarding chaining to generate the prediction result.

Combining the search method for fire suppression routes with ant colony algorithms and methods of analyzing twitter events on the highway is the basis of the problems to be studied. The results of the twitter data feature extraction are classified with Support Vector Machine after it is implemented in the Simple Additive Weighting method in calculating path weights with criteria of distance, congestion, multiple branching, and many holes. Line weights are also used as initial pheromone values. The C-means method is used to group the weights of each path and distance so that the path with the lowest weight and the shortest distance that will be simulated using the Ant Colony. The validation results with cross fold on SVM with linear kernels produce the greatest accuracy value is 97.93% for training data distribution: test data 6: 4. The simulation of the selection of the damkar car path from Feather to Pleburan with Ant Colony obtained 50 seconds of computation time, whereas with Ant Colony with Clustering the computation time was 15, resulting in a reduction in computing of 35. Ant colony with MinMax optimization gives the best computation time of 14.47 seconds with 100 iterations and 10 nodes.

In this interview with Ruben Hackler and Guido Kirsten, Franco Moretti elaborates on his biography and intellectual development, on 'distant reading' and his work at the Stanford Literary Lab, on the complicated relationships between literature, market, and ideology, and above all on the question what literary history might contribute to a social critique of the present. With regard to his experiences in the growing field of digital humanities, Moretti makes an argument for 'computational criticism' as a combination of data-driven research methods and in-depth analyses of literary texts.

This paper evaluates Franco Moretti's 'distant reading' from the perspective of Foucauldian discourse analysis and basic principles of computer science. After a short account of the essential features of both Moretti's and Foucault's approaches, the author assesses their combined potential for automation or algorithmization. This is accomplished by focusing on the decompositional and operational aspects of Foucauldian archeology, which seems to provide (at least in principle) a way to realize or implement Moretti's idea of an automated distant reading. The technical feasibility of this theoretical solution is eventually criticized in the light of the constraints and limitations of computation compared to 'human interfaces'.