Automated Classification to Improve the Efficiency of Weeding Library Collections

Previous studies have shown that weeding a library collection benefits patrons and increases circulation rates. However, the time required to review the collection and make weeding decisions presents a formidable obstacle. This study empirically evaluated methods for automatically classifying weeding candidates. A data set containing 80,346 items from a large-scale weeding project running from 2011 to 2014 at Wesleyan University was used to train six machine learning classifiers to predict a weeding decision of either ‘Keep’ or ‘Weed’ for each candidate. The study found statistically significant agreement (p?=?0.001) between classifier predictions and librarian judgments for all classifier types. The naive Bayes and linear support vector machine classifiers had the highest recall (fraction of items weeded by librarians that were identified by the algorithm), while the k-nearest-neighbor classifier had the highest precision (fraction of recommended candidates that librarians had chosen to weed). The variables found to be most relevant were: librarian and faculty votes for retention, item age, and the presence of copies in other libraries.     

The potential role of citizen conservation in re-shaping approaches to murals in an urban context

Public visual spaces, populated by a blend of community murals, unauthorised street art, and historic painted mercantile signs, are often the mark of an urban environment that is both progressive and eclectic. Changes in the aesthetic and cultural value of these urban mural forms have led to an increase in the appreciation and, in some instances, promotion of their artistic merit and cultural significance as examples of public art. However, examining the significance of these works, with a view to implementing a conservation approach is problematic. This is due to a number of practical and theoretical considerations that are primarily a result of the ephemeral existence of urban murals outside conventional exhibition spaces, and issues associated with their often fragmented ownership and uncertain authorship. Consequently, larger thinking on the interpretation, conservation assessment, and advocacy for the conservation of urban murals are required. Key to defining and implementing such strategies is contextualising the public visual spaces that these murals occupy and, as part of this, the local and wider communities’ perception of these murals as culturally significant objects as well as fostering awareness and understanding of appropriate measures aimed at their conservation. This paper examines the role of citizen science, or crowd-sourcing, of local community members in establishing a conservation dialogue and generating conservation- relevant data on urban murals. It looks specifically at a project involving a collection of in situ historic painted mercantile signs — also known as ghost signs — in the City of Port Phillip, Melbourne, Australia. The project fostered the establishment of an informed and open dialogue between conservation specialists and participants from the local community on the significance of local ghost signs whilst transferring knowledge on conservation processes and assessment methods. Working directly with community members, a programme was designed in which conservation and community knowledge of these urban art forms, could be collected and exchanged across digital platforms. This enabled researchers to examine how citizen science can be utilised as a research tool as well as a means to advocate for the conservation of collections of urban murals. It created the opportunity to consider the role of non-specialists and shared authorities in the collection and collation of conservation- relevant data and how information generated from what we call citizen conservation projects, can inform the way in which conservators evaluate and prioritize the conservation of urban cultural heritage. The data gathered and interpreted proved to be the most effective means of ‘conserving’ these often ephemeral forms of cultural material.     

张掖大佛寺取经壁画应是《西游记》的衍生物

张掖大佛寺的取经壁画被有关方面宣布是元代作品,早于吴承恩《西游记》200多年,是《西游记》的创作原型。这个结论值得怀疑。取经故事的形成经历了几个不同的发展阶段,各有特征可寻。从壁画内容看,其中的一些故事相当晚出,应是吴承恩《西游记》的衍生物。     

竹林七贤的几个基本问题

竹林七贤得名的缘由、过程,竹林七贤的性质和领袖等基本问题都是应重新探讨的问题。竹林"七贤"的由来不是《论语》"作者七人",而是汉末"七贤";竹林七贤的得名经历了从竹林之游、竹林名士到竹林七贤的历史过程;竹林七贤是在阴谋集团篡国夺权形势下知识分子避世隐居的集团,七贤的中心是嵇、阮、山三人,而领袖是嵇康。     

唐代墓室壁画艺术风格初探

唐代墓室壁画填补了唐画研究缺乏真迹的空白,展现了唐代绘画的创新面貌。已发现的唐代墓室壁画可分为三个发展时期初期继承南北朝画风,疏、密二体并行,由疏简、劲健而稍有板滞向流畅、舒展而又灵动的方向转化;中期体现为几种风格的融合,简快、豪放,有“吴带当风”之韵;晚期在完善中期风格的同时,又有创新,开五代细丽精巧的人物画风及工笔重彩花鸟画风之先河。     

亦真亦幻的佛国世界——浅议库木吐拉的汉风壁画

库木吐拉石窟位于库车县库木吐拉村,是新疆汉风壁画遗存最多的一处石窟。库木吐拉的汉风洞窟已可初步确认的达20多个,尚有许多因风沙和水渍的毁坏已无法统计。本文选择其中最具代表性的11、14、16、73号洞窟等,就其风格技法作初步的探讨。     

论麦积山石窟北朝壁画飞天造型特征及其来源

飞天在印度只是风筝型的形象,多呈现出一种静止感,然而伴随着佛教及佛教艺术的传播,飞天经中亚、沿着丝绸之路传入了中国后,开始借助衣裙和飘带,创造出韵律柔美的飞动气势和美丽姿态。本文以麦积山石窟北朝壁画为切入点,对其独有的飞天造型特征予以分析研究,推导出复杂的演变过程,并纳入其他相关的飞天造型进行比较,以探求其特征来源。     

在“敦煌莫高窟专题研讨会”上的总结发言

本文是作者在"敦煌莫高窟专题研讨会"上的总结发言。作为考古学者,他认为现有的考古证据表明,莫高窟在结构、功能上,与印度及中国新疆石窟有所不同,但是与中原豪强多墓室的墓葬壁画更为相似。在这一点上,作者赞同与会学者沙尔夫教授与索尼娅·李及汪悦进等人的观点:即便是非常明显的叙事性壁画,其彩绘装饰也不一定是说教性的;敦煌莫高窟的造像本质上是用来自我表现的,与汉代墓室中的壁画装饰非常相似。     

库木吐喇石窟第56窟空鼓及起甲壁画的抢救性保护修复

库木吐喇石窟第56窟为中心柱券顶窟,其壁画保存状况极差,所存壁画均被烟熏。壁画颜料层起甲后呈大片斑驳脱落,券顶两侧的壁画起甲尤甚,轮廓极其模糊。此外,地仗空鼓严重、大面积脱离岩体,时有掉落的危险。为此,通过起甲壁画加固、空鼓地仗层灌浆加固、风化砂岩支撑体加固等方法对第56窟壁画进行了抢救性的保护修复,解除了目前威胁壁画保存的病害,保证目前条件下的安全和稳定,使之能够长期保存下去。     

新疆建筑设施名称的翻译浅议

本文从范仲英的翻译三原则角度对新疆建筑设施名称的翻译进行了举例讨论,分别探讨了寺庙、洞窟和壁画以及墓葬的翻译,得出的结论是,在新疆的多元文化背景下,译文既要最大限度的忠实于读者,又要最大限度的忠实于名称的文化内涵。本文对新疆其他类别的旅游景点名称翻译有借鉴意义。