A Model for Recommending Related Research Papers: A Natural Language Processing Approach
- Authors: Van Heerden, Juandre Anton
- Date: 2022-04
- Subjects: Electronic information resources , Research
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/58495 , vital:59651
- Description: The volume of information generated lately has led to information overload, which has impacted researchers’ decision-making capabilities. Researchers have access to a variety of digital libraries to retrieve information. Digital libraries often offer access to a number of journal articles and books. Although digital libraries have search mechanisms it still takes much time to find related research papers. The main aim of this study was to develop a model that uses machine learning techniques to recommend related research papers. The conceptual model was informed by literature on recommender systems in other domains. Furthermore, a literature survey on machine learning techniques helped to identify candidate techniques that could be used. The model comprises four phases. These phases are completed twice, the first time for learning from the data and the second time when a recommendation is sought. The four phases are: (1) identify and remove stopwords, (2) stemming the data, (3) identify the topics for the model, and (4) measuring similarity between documents. The model is implemented and demonstrated using a prototype to recommend research papers using a natural language processing approach. The prototype underwent three iterations. The first iteration focused on understanding the problem domain by exploring how recommender systems and related techniques work. The second iteration focused on pre-processing techniques, topic modeling and similarity measures of two probability distributions. The third iteration focused on refining the prototype, and documenting the lessons learned throughout the process. Practical lessons were learned while finalising the model and constructing the prototype. These practical lessons should help to identify opportunities for future research. , Thesis (MA) -- Faculty of Engineering, the Built Environment, and Technology, 2022
- Full Text:
- Date Issued: 2022-04
- Authors: Van Heerden, Juandre Anton
- Date: 2022-04
- Subjects: Electronic information resources , Research
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/58495 , vital:59651
- Description: The volume of information generated lately has led to information overload, which has impacted researchers’ decision-making capabilities. Researchers have access to a variety of digital libraries to retrieve information. Digital libraries often offer access to a number of journal articles and books. Although digital libraries have search mechanisms it still takes much time to find related research papers. The main aim of this study was to develop a model that uses machine learning techniques to recommend related research papers. The conceptual model was informed by literature on recommender systems in other domains. Furthermore, a literature survey on machine learning techniques helped to identify candidate techniques that could be used. The model comprises four phases. These phases are completed twice, the first time for learning from the data and the second time when a recommendation is sought. The four phases are: (1) identify and remove stopwords, (2) stemming the data, (3) identify the topics for the model, and (4) measuring similarity between documents. The model is implemented and demonstrated using a prototype to recommend research papers using a natural language processing approach. The prototype underwent three iterations. The first iteration focused on understanding the problem domain by exploring how recommender systems and related techniques work. The second iteration focused on pre-processing techniques, topic modeling and similarity measures of two probability distributions. The third iteration focused on refining the prototype, and documenting the lessons learned throughout the process. Practical lessons were learned while finalising the model and constructing the prototype. These practical lessons should help to identify opportunities for future research. , Thesis (MA) -- Faculty of Engineering, the Built Environment, and Technology, 2022
- Full Text:
- Date Issued: 2022-04
Ict literacy skills and demographic factors as determinants of electronic resources use among the undergraduate students in the selected universities the Eastern Cape, South Africa
- Olatoye , Oluwayemi IbukunOluwa
- Authors: Olatoye , Oluwayemi IbukunOluwa
- Date: 2019
- Subjects: Electronic information resources
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10353/16176 , vital:40675
- Description: In today’s world, information is the foundation on which every strata in society is built and established. As we are in the jet age, the use of Information Communications Technology (ICT) is sine-qua-non to academic development. It is equally important to acquire skills and build capacity in ICT applications, as well as reflect on the demographic factors that determine the utilization of electronic resources among the undergraduate respondents. ICT has also evolutionalized professionalism in librarianship by providing delivery of appropriate, suitable and value-added information services in digital format. This research, therefore, investigated undergraduate students’ ICT literacy skills and demographic factors as determinants of electronic resources use, with selected tertiary institutions of learning in Eastern Cape South Africa as a case study. The study was premised on the Diffusion of Innovation Theory (DOI), Technology Acceptance Model (TAM) and the Theory of Reasoned Action (TRA) with the aim of appraising undergraduate students’ ICT literacy skills and demographic factors as causative elements of e-resources utilization in designated Eastern Cape universities in South Africa, as well as to unravel the impact of the theories on the adoption of technology and the perceived utilization of the electronic resources. The application of DOI, TAM and TRA theories for this study exemplifies the acceptance and usage of technological innovations by envisioned users in ICT literacy skill and electronic resources research, and these theories formed the theoretical basis to strengthen the study. The specific x objectives of the study are: To ascertain how undergraduate students in selected Higher Education Institutions (HEIs) in the Eastern Cape access e-resources; to determine the level of influence of ICT literacy skills on the use of electronic resources by Undergraduate students in the selected universities; to determine the regularity levels of use and problems encountered in the use of electronic resources by Undergraduate students in the selected universities; to ascertain the contributions of demographic factors on the use of electronic resources by Undergraduate students in the selected universities; and to determine the attitudes and perceptions of undergraduate students towards the use of eresources. The approach of the study was in threefold; one, general discussion regarding ICT literacy skills of the respondents and secondly the demographic factors that determine electronic resources use of undergraduate students in the University of Fort Hare and Rhodes University. Finally, ICT literacy skills and demographic factors were investigated with the applicability of TAM, DOI and TRA theories Specifically, under these theories (TAM, TRA and DOI), TAM and TRA models were used to explain behavioural intention and to envisage user acceptance of technology usage (electronic resources), and to elucidate the correlation between the respondent’s (undergraduate students) perceptions, attitudes, beliefs and ultimately system utilization. DOI was conceptualized in this study as a valued tool for appraising the effect of demographic factors on the utilization of electronic resources among the undergraduate students in their academic pursuit. The major findings of the study specifies that ICT literacy skills and demographic factors determine the use of electronic resources. Hence, it is reasoned in the thesis that ICT xi literacy and demographic factors affects the frequency of electronic resources with those, for instance, who have obtained high ICT literacy skill levels when compared to others who are yet to develop their ICT literacy skills. Further, it has been disclosed elsewhere in the study that in terms of age, the younger undergraduate students (from 21 to 30 years) utilize electronic resources more regularly than their older colleagues (those who are 30 years of age and above).The study was approached with the adoption of the mixed-method research technique. The administration of a total of 377 copies of the questionnaire to undergraduate respondents in the aforementioned HEIs, (out of which 266 copies were returned), was conducted with in-depth interview conversations comprising of ten participants, with six respondents selected in the University of Fort Hare and four respondents from Rhodes University. Data acquired from the study were processed and analyzed with the aid of Statistical Package for Social Science (SPSS) for the quantitative data. In the light of the theoretical frameworks of the study, research results established that the ICT experience of the undergraduate respondents greatly influences their proficiency levels. This hypothesized assertion was subjected to statistical validity test through regression analysis. The result depicts that the p-value is 0.49 (which means that p≤ 0.05), and interprets to mean that the hypothesis is accepted. Also, the findings of this study depicts that the utilization of electronic resources by the respondents is mostly for entertainment purposes (such as viewing online videos, listening to sport commentaries, music and video downloads, e-mail communications, chatting with other people) had the highest rankings from the component matrix analysis which were greater than 0.5. From the forgoing, this is interpreted to mean that the respondents possess excellent proficiency in ICT literacy skills as well as in the use of Microsoft packages. xii Also, in the course of the in-depth research interview, it was discovered that most of the interviewees have excellent proficiency in ICT literacy skills. Generally, gender is an essential element that determines accessibility and e-resource utilization of respondents to electronic resources through the home and from other sources. Furthermore, it was discovered that that language is not a determinant regarding respondents’ accessibility and e-resource utilization from other sources of access to respondents. The analysis of this study revealed that more males, who are within the active e-resource using age bracket of 21 to 30 years old access and utilize electronic resources through the residences than their female counterparts. This age bracket is followed in terms of access and use of e-resources through residences by the respondents that are 20 years and below. A chi-square test of independence was also performed to survey the level of correlation between age and access to E-resources. A small p-value (typically ≤ 0.05) indicates strong evidence against the null hypothesis, so you reject the null hypothesis. A large p-value (> 0.05) shows weak evidence against the null hypothesis, so you reject the null hypothesis i.e. X2 (3, N=53) = 7.82. The Pearson chi-square (p-value) generated was .294, which is construed to mean that it is insignificant. Therefore, the explanation is that age has no influence on access of respondents to electronic resources through cybercafé. In order to make ICT literacy skills more beneficial to the undergraduate students in the selected HEIs, recommendations were made in this study. Firstly, there is a need for mass enlightenment campaigns on the use and benefits of E-resources among undergraduate respondents, the building of capacity of the undergraduate students in the use of electronic resources ICT literacy skill development programmes, need for intervention programmes focusing on the application xiii of some E-resources and software where the students are ranked low. Further, it is recommended that female students need to be encouraged to use E-resources. Also, delivery and empowering of Wi-Fi services, as well as the provision of CD- ROM databases should be considered.
- Full Text:
- Date Issued: 2019
- Authors: Olatoye , Oluwayemi IbukunOluwa
- Date: 2019
- Subjects: Electronic information resources
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10353/16176 , vital:40675
- Description: In today’s world, information is the foundation on which every strata in society is built and established. As we are in the jet age, the use of Information Communications Technology (ICT) is sine-qua-non to academic development. It is equally important to acquire skills and build capacity in ICT applications, as well as reflect on the demographic factors that determine the utilization of electronic resources among the undergraduate respondents. ICT has also evolutionalized professionalism in librarianship by providing delivery of appropriate, suitable and value-added information services in digital format. This research, therefore, investigated undergraduate students’ ICT literacy skills and demographic factors as determinants of electronic resources use, with selected tertiary institutions of learning in Eastern Cape South Africa as a case study. The study was premised on the Diffusion of Innovation Theory (DOI), Technology Acceptance Model (TAM) and the Theory of Reasoned Action (TRA) with the aim of appraising undergraduate students’ ICT literacy skills and demographic factors as causative elements of e-resources utilization in designated Eastern Cape universities in South Africa, as well as to unravel the impact of the theories on the adoption of technology and the perceived utilization of the electronic resources. The application of DOI, TAM and TRA theories for this study exemplifies the acceptance and usage of technological innovations by envisioned users in ICT literacy skill and electronic resources research, and these theories formed the theoretical basis to strengthen the study. The specific x objectives of the study are: To ascertain how undergraduate students in selected Higher Education Institutions (HEIs) in the Eastern Cape access e-resources; to determine the level of influence of ICT literacy skills on the use of electronic resources by Undergraduate students in the selected universities; to determine the regularity levels of use and problems encountered in the use of electronic resources by Undergraduate students in the selected universities; to ascertain the contributions of demographic factors on the use of electronic resources by Undergraduate students in the selected universities; and to determine the attitudes and perceptions of undergraduate students towards the use of eresources. The approach of the study was in threefold; one, general discussion regarding ICT literacy skills of the respondents and secondly the demographic factors that determine electronic resources use of undergraduate students in the University of Fort Hare and Rhodes University. Finally, ICT literacy skills and demographic factors were investigated with the applicability of TAM, DOI and TRA theories Specifically, under these theories (TAM, TRA and DOI), TAM and TRA models were used to explain behavioural intention and to envisage user acceptance of technology usage (electronic resources), and to elucidate the correlation between the respondent’s (undergraduate students) perceptions, attitudes, beliefs and ultimately system utilization. DOI was conceptualized in this study as a valued tool for appraising the effect of demographic factors on the utilization of electronic resources among the undergraduate students in their academic pursuit. The major findings of the study specifies that ICT literacy skills and demographic factors determine the use of electronic resources. Hence, it is reasoned in the thesis that ICT xi literacy and demographic factors affects the frequency of electronic resources with those, for instance, who have obtained high ICT literacy skill levels when compared to others who are yet to develop their ICT literacy skills. Further, it has been disclosed elsewhere in the study that in terms of age, the younger undergraduate students (from 21 to 30 years) utilize electronic resources more regularly than their older colleagues (those who are 30 years of age and above).The study was approached with the adoption of the mixed-method research technique. The administration of a total of 377 copies of the questionnaire to undergraduate respondents in the aforementioned HEIs, (out of which 266 copies were returned), was conducted with in-depth interview conversations comprising of ten participants, with six respondents selected in the University of Fort Hare and four respondents from Rhodes University. Data acquired from the study were processed and analyzed with the aid of Statistical Package for Social Science (SPSS) for the quantitative data. In the light of the theoretical frameworks of the study, research results established that the ICT experience of the undergraduate respondents greatly influences their proficiency levels. This hypothesized assertion was subjected to statistical validity test through regression analysis. The result depicts that the p-value is 0.49 (which means that p≤ 0.05), and interprets to mean that the hypothesis is accepted. Also, the findings of this study depicts that the utilization of electronic resources by the respondents is mostly for entertainment purposes (such as viewing online videos, listening to sport commentaries, music and video downloads, e-mail communications, chatting with other people) had the highest rankings from the component matrix analysis which were greater than 0.5. From the forgoing, this is interpreted to mean that the respondents possess excellent proficiency in ICT literacy skills as well as in the use of Microsoft packages. xii Also, in the course of the in-depth research interview, it was discovered that most of the interviewees have excellent proficiency in ICT literacy skills. Generally, gender is an essential element that determines accessibility and e-resource utilization of respondents to electronic resources through the home and from other sources. Furthermore, it was discovered that that language is not a determinant regarding respondents’ accessibility and e-resource utilization from other sources of access to respondents. The analysis of this study revealed that more males, who are within the active e-resource using age bracket of 21 to 30 years old access and utilize electronic resources through the residences than their female counterparts. This age bracket is followed in terms of access and use of e-resources through residences by the respondents that are 20 years and below. A chi-square test of independence was also performed to survey the level of correlation between age and access to E-resources. A small p-value (typically ≤ 0.05) indicates strong evidence against the null hypothesis, so you reject the null hypothesis. A large p-value (> 0.05) shows weak evidence against the null hypothesis, so you reject the null hypothesis i.e. X2 (3, N=53) = 7.82. The Pearson chi-square (p-value) generated was .294, which is construed to mean that it is insignificant. Therefore, the explanation is that age has no influence on access of respondents to electronic resources through cybercafé. In order to make ICT literacy skills more beneficial to the undergraduate students in the selected HEIs, recommendations were made in this study. Firstly, there is a need for mass enlightenment campaigns on the use and benefits of E-resources among undergraduate respondents, the building of capacity of the undergraduate students in the use of electronic resources ICT literacy skill development programmes, need for intervention programmes focusing on the application xiii of some E-resources and software where the students are ranked low. Further, it is recommended that female students need to be encouraged to use E-resources. Also, delivery and empowering of Wi-Fi services, as well as the provision of CD- ROM databases should be considered.
- Full Text:
- Date Issued: 2019
The use of electronic information resources in the university of Fort Hare Library Services
- Authors: Maya, Zukiswa
- Date: 2018
- Subjects: Acquisition of electronic information resources , Electronic information resources , Collection management (Libraries)
- Language: English
- Type: Thesis , Masters , MLIS
- Identifier: http://hdl.handle.net/10353/6303 , vital:29557
- Description: The study seeks to explore the use of electronic information resource in the University of Fort Hare (UFH) Library. The objectives of the study are to determine factors that influence acquisitions of electronic information resources at UFH library, to find out the user’s responses to electronic information resources in the library and identify the challenges faced by UFH library regarding the usage of electronic information resources. The literature review was conducted through an acquisition of electronic information resources in academic libraries, collection development policies of academic libraries in South Africa and application of electronic information resources within South Africa and globally. The study is based on Diffusion of innovation (DOI) theory. The study adopted qualitative and quantitative approaches, and the non-probability sampling-Quota sampling was used for students and purposive sampling technique for librarians and academics. The data was collected with self-administered questionnaires and document analysis. The study found that academics were not fully involved in the acquisition of the library electronic information resources; therefore, there is a lack of communication about the acquisition of electronic resources. The study further reveals that there is usage of electronic information resources; however, there are library users who prefers to use search engines such as google, yahoo etc. It was also identified that there are two important barriers that hinder the use of electronic information resources, i.e. physical and personal barriers. The study recommends that University of Fort Hare library should consider including e resources in the collection development policy. It is also recommended that the library online training/tutorials must be installed on the library website to increase the usage of e-resources. In order to stay relevant and visible, librarians should embrace new opportunities and go beyond the comfort zone of traditional librarian principles.
- Full Text:
- Date Issued: 2018
- Authors: Maya, Zukiswa
- Date: 2018
- Subjects: Acquisition of electronic information resources , Electronic information resources , Collection management (Libraries)
- Language: English
- Type: Thesis , Masters , MLIS
- Identifier: http://hdl.handle.net/10353/6303 , vital:29557
- Description: The study seeks to explore the use of electronic information resource in the University of Fort Hare (UFH) Library. The objectives of the study are to determine factors that influence acquisitions of electronic information resources at UFH library, to find out the user’s responses to electronic information resources in the library and identify the challenges faced by UFH library regarding the usage of electronic information resources. The literature review was conducted through an acquisition of electronic information resources in academic libraries, collection development policies of academic libraries in South Africa and application of electronic information resources within South Africa and globally. The study is based on Diffusion of innovation (DOI) theory. The study adopted qualitative and quantitative approaches, and the non-probability sampling-Quota sampling was used for students and purposive sampling technique for librarians and academics. The data was collected with self-administered questionnaires and document analysis. The study found that academics were not fully involved in the acquisition of the library electronic information resources; therefore, there is a lack of communication about the acquisition of electronic resources. The study further reveals that there is usage of electronic information resources; however, there are library users who prefers to use search engines such as google, yahoo etc. It was also identified that there are two important barriers that hinder the use of electronic information resources, i.e. physical and personal barriers. The study recommends that University of Fort Hare library should consider including e resources in the collection development policy. It is also recommended that the library online training/tutorials must be installed on the library website to increase the usage of e-resources. In order to stay relevant and visible, librarians should embrace new opportunities and go beyond the comfort zone of traditional librarian principles.
- Full Text:
- Date Issued: 2018
A data warehouse structure design methodology to support the efficient and effective analysis of online resource usage data
- Authors: Ferreira, Cornél
- Date: 2012
- Subjects: Data warehousing , Electronic information resources
- Language: English
- Type: Thesis , Masters , MA
- Identifier: vital:10486 , http://hdl.handle.net/10948/d1016072
- Description: The use of electronic services results in the generation of vast amounts of Online Resource Usage (ORU) data. ORU data typically consists of user login, printing and executed process information. The structure of this type of data restricts the ability of decision makers to effectively and efficiently analyse ORU data. A data warehouse (DW) structure is required which satisfies an organisation’s information requirements. In order to design a DW structure a methodology is needed to provide a design template according to acknowledged practices. The aim of this research was to primarily propose a methodology specifically for the design of a DW structure to support the efficient and effective analysis of ORU data. A variety of relevant DW structure design methodologies were investigated and a number of limitations were identified. These methodologies do not provide methodological support for metadata documentation, physical design and implementation. The most comprehensive methodology identified in the investigation was modified and the Adapted Triple-Driven DW Structure Design Methodology (ATDM) was proposed. The ATDM was successfully applied to the information and communication technology services (ICTS) department of the Nelson Mandela Metropolitan University as the case study for this research. The proposed ATDM consists of different phases which include a requirements analysis phase that was adapted from the identified comprehensive methodology. A physical design and an implementation phase were included in the ATDM. The ATDM was successfully applied to the ICTS case study as a proof of concept. The application of the ATDM to ICTS resulted in the generation and documentation of semantic and technical metadata which describes the DW structure derived from the application of the ATDM at a logical and physical level respectively. The implementation phase was applied using the Microsoft SQL Server integrated tool to obtain an implemented DW structure for ICTS that is described by technical metadata at an implementation level. This research has shown that the ATDM can be successfully applied to obtain an effective and efficient DW structure for analysing ORU data. The ATDM provides guidelines to develop a DW structure for ORU data and future research includes the generalisation of the ATDM to accommodate various domains and different data types.
- Full Text:
- Date Issued: 2012
- Authors: Ferreira, Cornél
- Date: 2012
- Subjects: Data warehousing , Electronic information resources
- Language: English
- Type: Thesis , Masters , MA
- Identifier: vital:10486 , http://hdl.handle.net/10948/d1016072
- Description: The use of electronic services results in the generation of vast amounts of Online Resource Usage (ORU) data. ORU data typically consists of user login, printing and executed process information. The structure of this type of data restricts the ability of decision makers to effectively and efficiently analyse ORU data. A data warehouse (DW) structure is required which satisfies an organisation’s information requirements. In order to design a DW structure a methodology is needed to provide a design template according to acknowledged practices. The aim of this research was to primarily propose a methodology specifically for the design of a DW structure to support the efficient and effective analysis of ORU data. A variety of relevant DW structure design methodologies were investigated and a number of limitations were identified. These methodologies do not provide methodological support for metadata documentation, physical design and implementation. The most comprehensive methodology identified in the investigation was modified and the Adapted Triple-Driven DW Structure Design Methodology (ATDM) was proposed. The ATDM was successfully applied to the information and communication technology services (ICTS) department of the Nelson Mandela Metropolitan University as the case study for this research. The proposed ATDM consists of different phases which include a requirements analysis phase that was adapted from the identified comprehensive methodology. A physical design and an implementation phase were included in the ATDM. The ATDM was successfully applied to the ICTS case study as a proof of concept. The application of the ATDM to ICTS resulted in the generation and documentation of semantic and technical metadata which describes the DW structure derived from the application of the ATDM at a logical and physical level respectively. The implementation phase was applied using the Microsoft SQL Server integrated tool to obtain an implemented DW structure for ICTS that is described by technical metadata at an implementation level. This research has shown that the ATDM can be successfully applied to obtain an effective and efficient DW structure for analysing ORU data. The ATDM provides guidelines to develop a DW structure for ORU data and future research includes the generalisation of the ATDM to accommodate various domains and different data types.
- Full Text:
- Date Issued: 2012
- «
- ‹
- 1
- ›
- »