How Does AI Make Libraries Smart?: A Case Study of Hangzhou Public Library

Introduction

From the computer Deep Blue developed in 1997 to the computer program AlphaGo developed in 2016, board game battles between robots and humans constantly push Artificial Intelligence (AI) technology into the public’s view. Humans have mixed emotions about the victories of AI-based robots in games against humans. On the one hand, human beings are excited about the sea change that AI will bring to our lives. On the other hand, the possibilities of AI taking over most human beings’ work may leave humans with limited prospects for employment and economic growth. Due to the high-sensitive requirements of librarians as a profession and the fact that Chinese librarians have been in the shadow of the libraries declining for years, we paid close attention to the impact of AI on libraries.

The origin of AI is much earlier than the game robots, such as Deep Blue and AlphaGo. As early as 1950, Alan Turing, often referred to as the father of computer science (Beavers, 2013), started discussing machine thinking and artificial intelligence in his article Computing Machinery and Intelligence (Turning, 1950). The first proof of concept for Artificial Intelligence (AI) technology was demonstrated at John McCarthy’s first AI conference at Dartmouth college in 1956 (McCarthy et al., 2006). McCarthy and his colleagues stated that research about AI was to “proceed on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it” (McCarthy et al., 2006, p.2). It is related to the similar task of using computers to understand human intelligence, but AI does not have to confine itself to biologically observable methods (McCarthy, 2004).

In the half-century since the term AI was coined, the research and application of AI developed slowly, with ups and downs. Regardless of the symbolism genre, based on rules, symbols, algorithms, or the connectionism genre based on neural networks, it has encountered significant obstacles or even reached a dead end (Crevier, 1993; Nilsson, 1998; Russell & Norvig, 2002). What ultimately revived the neural network approach and AI was a change in the the explosive growth of sample data and computer capabilities, coupled with a significant technological breakthrough—deep learning (Li, 2018. Deep learning is a subset of machine learning (Guo et al., 2016), essentially a neural network with three or more layers (Rani & Kumar, 2019). These neural networks attempt to simulate the behavior of the human brain, allowing it to learn from large amounts of data (Bashar, 2019). While a neural network with only one layer can make approximate predictions, additional hidden layers help to optimize and refine for accuracy (IBM Cloud Education, 2020). Despite the success of deep learning research, AI systems are still in the early stage of narrow AI, indicating they can only handle single or limited tasks (Ben, 2017).

At present, machine learning, natural language processing, computer vision, and other technologies involved in AI are being integrated into various aspects of our lives (Meel, 2021; Section, 2020; Tyagi, 2021). In addition to bringing subversive changes to daily life fields such as speech recognition (The Signal, 2021), facial recognition (Klosowski, 2020), and machine translation (Madhavan, 2019), AI also shows excellent application prospects in various fields, such as medical treatment (Greenfield, 2019), transportation (Conde & Twinn, 2019), manufacturing (Manufacturer, 2021).