Artificial Neural Network to Improve Library Services

Artificial Neural Network to Improve Library Services

Artificial Neural Network (ANN) is a biologically inspired Machine Learning model that simulates how cells are connected in human brains to make a complicated decision. In topology terms, it can be represented by an acyclical network with explicit an input layer and an output layer which, respectively, doesn’t have to be organized in a structured manner. In mathematics, it manifests itself with an iterative application of perceptron function. The model can be used in a variety of ways, classification (the most common one), clustering, and language translation (usually achieved by networks of a specialized kind), just to name a few. Wayne State University explored the possibilities of using ANN to predict how likely a user was searching for one type of information. The poster will cover basics of ANN, on which the exploratory work was based, and the presenter will also share how to perform error back-propagation, which is not straightforwardly covered in most related literature.

Jiang, Minghao

2018 CALA Midwest Chapter Annual Program

2018-05-11

https://creativecommons.org/licenses/by/4.0/

https://midwest.chapters.cala-web.org/annual-programs/2018-cala-midwest-chapter-annual-conference/2018-cala-midwest-chapter-annual-program/

text

eng

Presentation

2018-05-11