Wyniki wyszukiwania - Biblioteka Nauki

1

Ontology-based recommendation in multimedia sharing systems

100%

Musiał K. , Juszczyszyn K. , Kazienko P.

Systems Science

|

2008

|

tom Vol. 34, no 1

97-106

EN

In this paper, a new framework for recommendation of multimedia objects in the environment of the multimedia sharing system has been proposed. It uses two kinds of individual ontologies, one is created for multimedia objects and the second one for system users. The final recommendation process takes into account similarities calculated both between objects' and users' ontologies. These individual ontologies respect all the social and semantic features existing in the system. The entire recommender framework was developed for the use in Flickr, a typical photo sharing system.

2

The recommendation algorithm for an online art gallery

80%

Karwowski W. , Sosnowska J. , Rusek M.

Information Systems in Management

|

2018

|

tom Vol. 7, No. 2

108--119

EN

The paper discusses the need for recommendations and the basic recommendation systems and algorithms. In the second part the design and implementation of the recommender system for online art gallery (photos, drawings, and paintings) is presented. The designed customized recommendation algorithm is based on collaborative filtering technique using the similarity between objects, improved by information from user profile. At the end conclusions of performed algorithm are formulated.

3

Application of multi-criteria analysis based on individual psychological profile for recommender systems

80%

Rafalak M. , Granat J. , Wierzbicki A. P.

Computer Science

|

2016

|

tom Vol. 17 (4)

503--517

EN

This paper presents a novel approach for user classification exploiting multi- criteria analysis. This method is based on measuring the distance between an observation and its respective Pareto front. The obtained results show that the combination of the standard KNN classification and the distance from Pareto fronts gives satisfactory classification accuracy – higher than the accuracy ob- tained for each of these methods applied separately. Conclusions from this study may be applied in recommender systems where the proposed method can be implemented as the part of the collaborative filtering algorithm.

4

Assessment Methods for Evaluation of Recommender Systems: A Survey

80%

Kuanr M. , Mohapatra P.

Foundations of Computing and Decision Sciences

|

2021

|

tom Vol. 46, No. 4

393--421

EN

The recommender system (RS) filters out important information from a large pool of dynamically generated information to set some important decisions in terms of some recommendations according to the user’s past behavior, preferences, and interests. A recommender system is the subclass of information filtering systems that can anticipate the needs of the user before the needs are recognized by the user in the near future. But an evaluation of the recommender system is an important factor as it involves the trust of the user in the system. Various incompatible assessment methods are used for the evaluation of recommender systems, but the proper evaluation of a recommender system needs a particular objective set by the recommender system. This paper surveys and organizes the concepts and definitions of various metrics to assess recommender systems. Also, this survey tries to find out the relationship between the assessment methods and their categorization by type.

5

Using cognitive models to understand and counteract the effect of self-induced bias on recommendation algorithms

80%

Pawłowska J. , Rydzewska K. , Wierzbicki A.

Journal of Artificial Intelligence and Soft Computing Research

|

2023

|

tom Vol. 13, No. 2

73--94

EN

Recommendation algorithms trained on a training set containing sub-optimal decisions may increase the likelihood of making more bad decisions in the future. We call this harmful effect self-induced bias, to emphasize that the bias is driven directly by the user’s past choices. In order to better understand the nature of self-induced bias of recommendation algorithms that are used by older adults with cognitive limitations, we have used agent-based simulation. Based on state-of-the-art results in psychology of aging and cognitive science, as well as our own empirical results, we have developed a cognitive model of an e-commerce client that incorporates cognitive decision-making abilities. We have evaluated the magnitude of self-induced bias by comparing results achieved by simulated agents with and without cognitive limitations due to age. We have also proposed new recommendation algorithms designed to counteract self-induced bias. The algorithms take into account user preferences and cognitive abilities relevant to decision making. To evaluate the algorithms, we have introduced 3 benchmarks: a simple product filtering method and two types of widely used recommendation algorithms: Content-Based and Collaborative filtering. Results indicate that the new algorithms outperform benchmarks both in terms of increasing the utility of simulated agents (both old and young), and in reducing self-induced bias.

6

Clustering Algorithms in Hybrid Recommender System on MovieLens Data

80%

Kuzelewska U.

Studies in Logic, Grammar and Rhetoric

|

2014

|

tom 37

|

nr 1

125-139

EN

Decisions are taken by humans very often during professional as well as leisure activities. It is particularly evident during surfing the Internet: selecting web sites to explore, choosing needed information in search engine results or deciding which product to buy in an on-line store. Recommender systems are electronic applications, the aim of which is to support humans in this decision making process. They are widely used in many applications: adaptive WWW servers, e-learning, music and video preferences, internet stores etc. In on-line solutions, such as e-shops or libraries, the aim of recommendations is to show customers the products which they are probably interested in. As input data the following are taken: shopping basket archives, ratings of the products or servers log files. The article presents a solution of recommender system which helps users to select an interesting product. The system analyses data from other customers' ratings of the products. It uses clustering methods to find similarities among the users and proposed techniques to identify users' profiles. The system was implemented in Apache Mahout environment and tested on a movie database. Selected similarity measures are based on: Euclidean distance, cosine as well as correlation coefficient and loglikehood function.

7

A new approach to image-based recommender systems with the application of heatmaps maps

80%

Woldan P. , Duda P. , Cader A. , Laktionov I.

Journal of Artificial Intelligence and Soft Computing Research

|

2023

|

tom Vol. 13, No. 2

63--72

EN

One of the fundamental issues of modern society is access to interesting and useful content. As the amount of available content increases, this task becomes more and more challenging. Our needs are not always formulated in words; sometimes we have to use complex data types like images. In this paper, we consider the three approaches to creating recommender systems based on image data. The proposed systems are evaluated on a real-world dataset. Two case studies are presented. The first one presents the case of an item with many similar objects in a database, and the second one with only a few similar items

8

Measuring Semantic Closeness of Ontologically Demarcated Resources

70%

Rhee S. K. , Lee J. , Park M.-W. , Szymczak M. , Frąckowiak G. , Ganzha M. , Paprzycki M.

Fundamenta Informaticae

|

2009

|

tom Vol. 96, nr 4

395-418

EN

In our work, an agent-based system supporting workers in an organization is centered around utilization of ontologically demarcated data. In this system, ontological matchmaking, understood as a way of establishing closeness between resources, is one of key functionalities. Specifically, it is used to autonomously provide recommendations to the user, who is represented by her/his personal agent. These recommendations specify, which among available resources are relevant / of interest to the worker. In this paper, we discuss our approach to measuring semantic closeness between ontologically demarcated information objects, while a Duty Trip Support application is used as a case study. General description of the algorithm is followed by a recommendation example based on support for a worker who is seeking advice in planning a duty trip.

9

ReSySTER: A hybrid recommender system for Scrum team roles based on fuzzy and rough sets

70%

Colomo-Palacios R. , González-Carrasco I. , López-Cuadrado J. , García-Crespo Á.

|

nr 4

801-816

EN

Agile development is a crucial issue within software engineering because one of the goals of any project leader is to increase the speed and flexibility in the development of new commercial products. In this sense, project managers must find the best resource configuration for each of the work packages necessary for the management of software development processes in order to keep the team motivated and committed to the project and to improve productivity and quality. This paper presents ReSySTER, a hybrid recommender system based on fuzzy logic, rough set theory and semantic technologies, aimed at helping project leaders to manage software development projects. The proposed system provides a powerful tool for project managers supporting the development process in Scrum environments and helping to form the most suitable team for different work packages. The system has been evaluated in a real scenario of development with the Scrum framework obtaining promising results.

10

Mining indirect association rules for web recommendation

70%

Kazienko P.

International Journal of Applied Mathematics and Computer Science

|

2009

|

tom 19

|

nr 1

165-186

EN

Classical association rules, here called “direct”, reflect relationships existing between items that relatively often co-occur in common transactions. In the web domain, items correspond to pages and transactions to user sessions. The main idea of the new approach presented is to discover indirect associations existing between pages that rarely occur together but there are other, “third” pages, called transitive, with which they appear relatively frequently. Two types of indirect associations rules are described in the paper: partial indirect associations and complete ones. The former respect single transitive pages, while the latter cover all existing transitive pages. The presented IDARM* Algorithm extracts complete indirect association rules with their important measure-confidence-using pre-calculated direct rules. Both direct and indirect rules are joined into one set of complex association rules, which may be used for the recommendation of web pages. Performed experiments revealed the usefulness of indirect rules for the extension of a typical recommendation list. They also deliver new knowledge not available to direct ones. The relation between ranking lists created on the basis of direct association rules as well as hyperlinks existing on web pages is also examined.

11

Resource optimisation in cloud computing : comparative study of algorithms applied to recommendations in a big data analysis architecture

70%

Ndayikengurukiye A. , Ez-Zahout A. , Aboubakr A. , Charkaoui Y. , Fouzia O.

Journal of Automation Mobile Robotics and Intelligent Systems

|

2021

|

tom Vol. 15, No. 4

65--75

EN

Recommender systems (RS) have emerged as a means of providing relevant content to users, whether in social networking, health, education, or elections. Furthermore, with the rapid development of cloud computing, Big Data, and the Internet of Things (IoT), the component of all this is that elections are controlled by open and accountable, neutral, and autonomous election management bodies. The use of technology in voting procedures can make them faster, more efficient, and less susceptible to security breaches. Technology can ensure the security of every vote, better and faster automatic counting and tallying, and much greater accuracy. The election data were combined by different websites and applications. In addition, it was interpreted using many recommendation algorithms such as Machine Learning Algorithms, Vector Representation Algorithms, Latent Factor Model Algorithms, and Neighbourhood Methods and shared with the election management bodies to provide appropriate recommendations. In this paper, we conduct a comparative study of the algorithms applied in the recommendations of Big Data architectures. The results show us that the K-NN model works best with an accuracy of 96%. In addition, we provided the best recommendation system is the hybrid recommendation combined by content-based filtering and collaborative filtering uses similarities between users and items.

12

On explainable fuzzy recommenders and their performance evaluation

60%

Rutkowski T. , Łapa K. , Nielek R.

International Journal of Applied Mathematics and Computer Science

|

2019

|

tom Vol. 29, no. 3

595--610

EN

This paper presents a novel approach to the design of explainable recommender systems. It is based on the Wang–Mendel algorithm of fuzzy rule generation. A method for the learning and reduction of the fuzzy recommender is proposed along with feature encoding. Three criteria, including the Akaike information criterion, are used for evaluating an optimal balance between recommender accuracy and interpretability. Simulation results verify the effectiveness of the presented recommender system and illustrate its performance on the MovieLens 10M dataset.

13

Transformation of "user-object" matrix for the collaborative filtering

51%

Savchuk T. , Sakaliuk A. , Wójcik W. , Burlibay A.

Przegląd Elektrotechniczny

|

2014

|

tom R. 90, nr 1

55-59

EN

The paper is devoted to application of collaborative filtering that is one of the method of automatic data filtering in the Internet. The main disadvantage of the approach is the necessity of performing a large number of operations. The authors have presented a mean of overcoming this problem by reduction of the dimension of the input matrix. Experimental results show that it had led not only to reduction of computational time, but also increased the accuracy of recommendations obtained.

PL

Artykuł poświęcony jest filtrowaniu kolaboracyjnemu, które jest jedną z metod automatycznej filtracji danych w sieci Internet. Główną wadą wspomnianego podejścia jest konieczność wykonywania bardzo dużej liczby operacji. Autorzy przedstawili rozwiązanie tego problemu polegający na redukcji wymiarowości przetwarzanej macierzy. Rezultaty badań pokazują, że oprócz zmniejszenia czasu obliczeń, uzyskano poprawę dokładności uzyskiwanych rekomendacji.

14

ReSySTER: A hybrid recommender system for Scrum team roles based on fuzzy and rough sets

51%

Colomo-Palacios R. , González-Carrasco I. , López-Cuadrado J. L. , García-Crespo Á.

International Journal of Applied Mathematics and Computer Science

|

2012

|

tom Vol. 22, no. 4

801-816

EN

Agile development is a crucial issue within software engineering because one of the goals of any project leader is to increase the speed and flexibility in the development of new commercial products. In this sense, project managers must find the best resource configuration for each of the work packages necessary for the management of software development processes in order to keep the team motivated and committed to the project and to improve productivity and quality. This paper presents ReSySTER, a hybrid recommender system based on fuzzy logic, rough set theory and semantic technologies, aimed at helping project leaders to manage software development projects. The proposed system provides a powerful tool for project managers supporting the development process in Scrum environments and helping to form the most suitable team for different work packages. The system has been evaluated in a real scenario of development with the Scrum framework obtaining promising results.

15

Clusterization methods for multi-variant e-commerce interfaces

51%

Wasilewski A.

Annals of Computer Science and Information Systems

|

2023

|

tom Vol. 35

309--313

EN

E-commerce is a very popular method that let consumers to purchase goods and services. The ability to purchase items online has increased the need for effective recommendation systems. Such recommendations relate usually to products in which the customer may be interested. However, there are wider opportunities to tailor e-commerce to individual customer needs and behaviour. In this paper the architecture of the ecommerce platform (named AIM2), which allows to provide a dedicated interface to selected user groups is discussed. A key component of the platform is the module responsible for dividing customers into groups, using selected clustering methods. Each of the implemented methods can be parameterised to adapt the customer segmentation to the requirements of the e-commerce owner. This article describes the results of an analysis of the impact of selected methods and parameters on clustering results. Moreover, it identifies key metrics that should be considered when selecting clustering conditions during the implementation of the platform. Finally, the main results of the pilot implementation of AIM2 are presented to assess the effectiveness of the multivariant user interface.

16

Mining indirect association rules for web recommendation

51%

Kazienko P.

International Journal of Applied Mathematics and Computer Science

|

2009

|

tom Vol. 19, no 1

165-186

EN

Classical association rules, here called 'direct', reflect relationships existing between items that relatively often co-occur in common transactions. In the web domain, items correspond to pages and transactions to user sessions. The main idea of the new approach presented is to discover indirect associations existing between pages that rarely occur together but there are other, 'third' pages, called transitive, with which they appear relatively frequently. Two types of indirect associations rules are described in the paper: partial indirect associations and complete ones. The former respect single transitive pages, while the latter cover all existing transitive pages. The presented IDARM* Algorithm extracts complete indirect association rules with their important measure-confidence-using pre-calculated direct rules. Both direct and indirect rules are joined into one set of complex association rules, which may be used for the recommendation of web pages. Performed experiments revealed the usefulness of indirect rules for the extension of a typical recommendation list. They also deliver new knowledge not available to direct ones. The relation between ranking lists created on the basis of direct association rules as well as hyperlinks existing on web pages is also examined.

17

Machine Learning as a method of adapting offers to the clients

51%

Bielecki J. , Ceglarski O. , Skublewska-Paszkowska M.

Journal of Computer Sciences Institute

|

2019

|

tom Vol. 13

267--271

EN

Recommendation systems are class of information filter applications whose main goal is to provide personalized recommendations. The main goal of the research was to compare two ways of creating personalized recommendations. The recommendation system was built on the basis of a content-based cognitive filtering method and on the basis of a collaborative filtering method based on user ratings. The conclusions of the research show the advantages and disadvantages of both methods.

PL

Systemy rekomendacji to aplikacje filtrujące dane, których głównym zadaniem jest dostarczanie spersonalizowanych rekomendacji produktów. Celem badań było dokonanie analizy i porównania dwóch metod uczenia maszynowego wykorzystywanych do generowania rekomendacji. System rekomendacji zbudowano na podstawie metody filtrowania kognitywnego opartej o treści oraz na podstawie metody filtrowania kolaboratywnego opartej o oceny użytkowników. Wnioski z przeprowadzonych badań pokazują wady i zalety obu metod.

18

Recommender system for navigation safety: requirements and methodology

51%

Shilov N.

TransNav : International Journal on Marine Navigation and Safety of Sea Transportation

|

2020

|

tom Vol. 14, no. 2

405--410

EN

Low maneuverability of ships together with growing intensity of marine traffic result in new challenges related to navigation safety. This paper reports a research aimed at design of methodology of operation of recommender systems for navigation safety. First, a specification of requirements to systems of the considered class has been carried out. Based on these, the major principles of functioning of such systems have been defined. The principles were a basis for development of the mentioned above methodology, which is based on the usage of context patterns and characterized by the presence of feedback to update the system’s knowledge base.

19

The Architecture of the Intelligent Case-Based Reasoning Recommender System (CBR RS) Recommending Preventive/Corrective Procedures in the Occupational Health and Safety Management System in an Enterprise

51%

Andreasik J.

Barometr Regionalny. Analizy i prognozy

|

2017

|

nr 3(49)

109-124

EN

The paper presents the original architecture of the system recommending preventive/corrective procedures in the occupational health and safety management system in an enterprise: ComplianceOHS-CBR. The system consists of four modules: Module A — an ontology of the workplace OHS profile, Module B — an ontology of preventive/corrective procedure indexation OPCPI, Module C — a recording system of the monitoring process of non-compliance with the requirements of OHS, Module D — a recommending engine consistent with the CBR methodology. The essence of the approach presented in this paper is integration of the monitoring system of the analysis process of non-compliance with the requirements of OHS at the workplace (the ADONIS system was used) with the case-based reasoning process (CBR). The integration platform consists of two ontologies: an ontology of profile compliance with the workplace OHS requirements (OP-OHS) and an ontology of preventive/corrective procedure indexation (OPCPI). Both of the ontologies are presented in the Protege 5 OWL editor. Inference engines are alternatively, according to the CBR methodology, myCBR and jCOLLIBRI.

PL

W pracy przedstawiono oryginalną architekturę systemu rekomendującego procedury zapobiegawczo-korygujące w systemie BHP przedsiębiorstwa: Compliance OHS-CBR. System składa się z czterech modułów: moduł A: ontologia profilu BHP stanowiska pracy, moduł B: ontologia indeksacji procedur zapobiegawczo-korygujących OIP-ZK, moduł C: system ewidencjonowania procesu monitorowania niezgodności z wymaganiami BHP, moduł D: silnik wydawania rekomendacji w metodologii CBR. Istotą podejścia prezentowanego w niniejszej pracy jest integracja systemu monitorowania procesu analizy niezgodności z wymaganiami BHP na stanowiskach pracy (zastosowano oprogramowanie ADONIS) z systemem wnioskowania z bazy przypadków CBR. Platformą integracji są dwie ontologie: ontologia profilu zgodności z wymaganiami BHP na stanowisku pracy (OP-BHP) oraz ontologia indeksacji procedur zapobiegawczo--korygujących OIP-ZK. Obydwie ontologie przedstawiono w edytorze Protege 5 języka OWL. Silnikami wnioskującymi zgodnie z metodologią CBR są alternatywnie: myCBR oraz jCOLLIBRI.

20

Improving Re-rankCCP with rules quality measures

41%

Jezusek P. , Karpus A.

Annals of Computer Science and Information Systems

|

2022

|

tom Vol. 30

61--71

EN

Recommender Systems are software tools and techniques which aim at suggesting new items that may possibly be of interest to a user. Context-Aware Recommender Systems exploit contextual information to provide more adequate recommendations. In this paper we described a modification of an existing contextual post-filtering algorithm which uses rules-like user representation called Contextual Conditional Preferences. We extended the algorithm by taking into account rules quality measures while recommending items to a user. We proved that this modification increases the quality of recommendations, measured with precision, recall and nDCG, and has no impact on the execution time of the original algorithm.