Browsing by Author "Gao, Wenli"
Now showing 1 - 15 of 15
- Results Per Page
- Sort Options
Item Analyzing data consultations: What liaisons can learn about users' data needs and use of tools(2017) Gao, Wenli; Ke, Irene; Martin, LisaAs data services gain momentum in academic libraries, liaison librarians are expected to be part of the education force for data literacy. Nonetheless, due to the broad scope of data literacy, training librarians to improve data-related skills can be challenging. This article used consultation statistics to identify tools and resources employed to solve users’ data questions. The results offered insights into users’ data needs and provided librarians with a clear direction to further develop data skills in their assigned disciplines. The methodology used in this study can be replicated at other institutions to identify needs and to direct professional development.Item Analyzing monograph usage of approval and firm orders for collection development(2015) Ke, Irene; Gao, Wenli; Bronicki, JacquelineThe poster describes a collection assessment project that compares the use of monographs purchased on an approval plan with those acquired via subject librarians’ selections. The goal is to reveal the usage trends between the two different selection processes. The analysis is based on LC classifications and subject groupings. We hypothesize that the usage patterns vary among subjects. The findings will inform collection managers in their effort to establish an effective and sustainable collection strategy.Item Beyond journal impact and usage statistics: Using citation analysis for collection development(2015) Gao, WenliAcademic libraries have a long history of assessing the collection and determining if the collection meets the need of academic departments. Citation analysis, a research method to understand users’ information behavior, allows the library to carry out this assessment. However, conducting a citation analysis is laborious. How can we employ current tools to make this time-consuming process productive? What kind of data can we draw from the study to inform collection development practice and articulate the value of library? This session describes a citation analysis project that examines citations in publications by communication faculty between 2006 and 2014 using SCOPUS. Citation analysis provides valuable information on material type, age, subject area, local core journals and titles for future purchase. This study also aims at finding if faculty is using and citing high impact journals. A Spearman’s correlation was performed to determine the relationship between journal citation count, journal impact factor (JCR), SCImago journal rank (SJR), and journal usage statistics. The preliminary finding of this study indicates that high impact journals were used more, but not being cited more. The session will detail data collection and analysis procedures and share the advantage and disadvantage of using SCOPUS as a citation analysis tool. It will also discuss the benefit of using SPSS to run analyses. Through discussion with participants about the value of conducting a citation analysis, the presenter hopes to spark interest among librarians in analyzing faculty citation behaviors as one way to evaluate library collection and using evidence-based practice to prove the value of library.Item Connecting the dots: Reader ratings, bibliographic data, and machine-learning algorithms for monograph selection(2019) Xiao, Jingshan; Gao, WenliTraditional collection development relies heavily on human input, with librarians relying on reviews and subject selection lists, and through user requests. With the development of machine learning, more and more businesses seek automated methods to deliver results relevant to users. The Recommender system, a subclass of information filtering that seeks to predict the "rating" or "preference" of a user, is among the most successful systems of machine learning in action. It has been adopted by many major e-commerce businesses such as Amazon, Netflix, and Expedia, and has been widely implemented to predict product and media recommendations, making it a key factor in increasing product average order value and the number of items per order. Drawing inspiration from the benefits of a recommender system to business and its success in heightening the reliability of recommendations, we attempted to build optimal collection recommendations with machine-learning algorithms using Python. The purpose of this project is to help librarians make collection decisions using the recommender system, and in this presentation we will illustrate several examples of building this system to aid in the selection of monographs. One example involves the merging of popular titles with reader rating data. We found that while The New York Times publishes best seller titles based on the rates of sales, they do not have any connection to user ratings. By leveraging data from Goodreads, the world’s largest site for readers and book recommendations, we will build a simple recommender system that produces The New York Times best seller titles that have higher user rating using a matrix factorization based method. Another example of using a recommender system is to have the ability to refer selectors to books that are similar to a particular title based on pairwise similarity scores. News services are already able to identify related articles of interest to readers based on the articles that they have read in the past, so applying this system to libraries is an exciting prospect. Drawing on bibliographic data from highly circulated items, the recommender system will suggest items with similar features using similarity metrics. The recommender system will use machine-learning algorithms not only to simplify collection development for librarians, but will also help end users discover more items relevant to their interests.Item Data mining, visualizing, and analyzing faculty thematic relationships for research support and collection analysis(2017) Gao, Wenli; Wallace, LorettaThis project developed analytical tools to improve collection development and enhance research support. Analyzing faculty publications across campus involves developing similarity analysis, topic generating algorithms, and visualization techniques. Using these tools, librarians can monitor research trends for collection development, and also enable customized research support for targeted faculty members. Besides, faculty can use the interactive Tableau visualization to find potential collaborators among their peers and boost collaboration at the university.Item Data visualization: New opportunities for outreach in the social sciences.(2015) Been, Joshua; Gao, Wenli; Wallace, LorettaThis presentation showcases several examples of using data visualization for outreach.Item Data visualization: Transforming research support(2016) Been, Joshua; Gao, Wenli; Wallace, LorettaThis presentation showcases how data visualization services can transform research support, from solely providing access to data, to actively engaging researchers with data. We will highlight Tableau, ESRI desktop and cloud-based applications. Examples will include energy research, consumer demographic analyses, and social media visualizations.Item Making evidence-based collection development feasible: Using R coding to automate analysis for action(2016) Gao, Wenli; Turner, Cherie; Ke, IreneCollection development practitioners face a complicated publishing environment with many new and developing purchasing models. Even with the new models in development, approval plans and firms orders are still commonly used in many academic libraries. In the literature, many have demonstrated ways of gathering monograph usage to inform collection development decisions. Nonetheless, we have rarely seen studies that have demonstrated ways to incorporate study results to form a detailed action plan and change collection practice. This research takes monograph usage study to the next level. Instead of analyzing the usage patterns of monograph collection as a whole, we compare the usage patterns of books acquired via approval plan and firm orders and incorporate interlibrary loan data for analysis. In order to make the work manageable and sustainable, an R script was developed to automate the analysis process. The results highlight call number ranges in granular levels where purchasing changes may be needed, allowing the selector to quickly identify the areas that need attention. The data provided by the script, in combination with a review by the selector, can provide the information needed to make effective changes to approval plans and firm order practice. In this presentation, we will discuss the rationale behind conducting this analysis, show how coding can make the analysis manageable and sustainable, and demonstrate the impact of this analysis on collection practices in various subject areas.Item Mapping the undergraduate curriculum for information literacy outreach and instruction in communication(2015) Gao, WenliThis session demonstrates a methodology to help librarians identify targets for information literacy outreach. We want to target high impact courses with information literacy requirements that are taken by a large number of communication and psychology majors. This methodology can be applied to other disciplines.Item A picture is worth a thousand words: Data visualization for collection assessment(2016-06-26) Been, Joshua; Gao, Wenli; Wiersma, GabrielleWhile a number of presentations and webinars have talked about data visualization for library assessment in general, there have been few such presentations focusing on data visualization for collection assessment data. This presentation will focus on using data visualizations tools to tell meaningful stories about a number of library collection assessment projects. We will demonstrate how to visually analyze collection data by creating interactive dashboards using Tableau software. Examples of visualizations include 1) comparing usage of print monographs purchased through approval plan and discretionary order, 2) serials cancellation data, and 3) correlations among faculty citation, electronic usage, and impact factors. By using visualizations, librarians can better understand their data, and more effectively communicate stories with internal and external stakeholders. In addition to showcasing collection analysis visualizations, we will discuss practical steps librarians can take to include visualization techniques in their analysis. We also created and will share a template using Tableau that other libraries can use to duplicate our visualizations.Item Piloting emerging research workshops at University of Houston(2018-11) Gao, WenliUniversity of Houston has recently launched a new initiative called 50 in 5, which aims to increase research output by 50% in 5 years. The liaison department has restructured to have a research team to better respond to the new research needs from faculty and students. As the data librarian, I lead our new emerging research workshop series to better equip faculty and students with new research tools for data cleaning, visualization, and GIS. With limited resources, one of my goals is also to provide training for librarians so that they can answer data related questions, and take on introductory level workshops so that I could develop more advanced courses. In this workshop, I will talk about how I design and promote the courses, feedback from participants, and plans for future courses. I will also talk about resources for librarians to develop skills in those emerging research areas. At the end of the session, participants will have an understanding of our pilot workshops and know how we design the courses; Participants will learn how to build a team of librarians in order to sustain and expand workshop series; Participants will be introduced to resources if they want to learn more in those emerging research tools.Item Transform research and instruction support through social media analysis and data visualization.(2017) Gao, WenliThis presentation showcases how social media analysis and data visualization can transform research and instruction support, from solely providing access to data, to actively engaging users with data. Participants will understand how to integrate NodeXL and Tableau in the classroom, and discuss costs, licensing, workflows, training considerations, and lessons learned.Item Using easily available data to build collections and understand faculty research for subject librarians(2018) Gao, WenliBuilding collection and assisting with faculty research are two major responsibilities for subject librarians. Collection and research needs are unique to campuses and departments subject librarians serve for. How could subject librarians use data to inform their collection decisions and understand their faculty research without spending significant amount of time collecting data? This presentation will show what you can do with data that have already been collected or easily retrieved. Usage data, interlibrary loan data, faculty publication and citation data were used in four projects to help the librarian uncover the collection and research needs for school of communication at University of Houston. This presentation will focus on the research process so that the methodology can be applied to other subject disciplines.Item What the Libraries do for your data needs: A conversation with UH Libraries research services(2020-07-17) Gao, Wenli; Malone, AndreaThe University of Houston (UH) Libraries is building programs and data-related services to support research that creates and utilizes large amounts of data. In this session, we will discuss the resources and services we provide and share examples of how we have worked with graduate students. We also want to learn what data needs you encounter so that we could tailor our services to fit your needs.Item You may also like: Machine-learning algorithms for collection recommendations(2019) Xiao, Jingshan; Gao, WenliTraditional collection development relies heavily on human input, with librarians relying on reviews and subject selection lists, and through user requests. With the development of machine learning, more and more businesses seek automated methods to deliver results relevant to users. The Recommender system, a subclass of information filtering that seeks to predict the "rating" or "preference" of a user, is among the most successful systems of machine learning in action. It has been adopted by many major e-commerce businesses such as Amazon, Netflix, and Expedia, and has been widely implemented to predict product and media recommendations, making it a key factor in increasing product average order value and the number of items per order. Drawing inspiration from the benefits of a recommender system to business and its success in heightening the reliability of recommendations, we attempted to build optimal collection recommendations with machine-learning algorithms using Python. The purpose of this project is to help librarians make collection decisions using the recommender system, and in this presentation we will illustrate several examples that will appeal to both public and academic librarians. One example involves the merging of popular titles with reader rating data. We found that while The New York Times publishes best seller titles based on the rates of sales, they do not have any connection to user ratings. By leveraging data from Goodreads, the world’s largest site for readers and book recommendations, we will build a simple recommender system that produces The New York Times best seller titles that have higher user rating using a matrix factorization based method. Librarians in turn can purchase best seller titles with good reader ratings. Moreover, the recommender system will also enable librarians to recommend books that are similar to a particular title based on pairwise similarity scores. Therefore, if a user enjoys reading a book, the recommender system will pull titles with similar features. The recommender system can be used in other settings. Drawing on bibliographic data from highly circulated items, or frequently requested interlibrary loan items, the recommender system will suggest items with similar features using similarity metrics. As a result, librarians can acquire relevant materials based on users’ previous reading patterns. The recommender system will use machine-learning algorithms not only to simplify collection development for librarians, but will also help end users discover more items relevant to their interests.