Construction Of High School History Knowledge Base For Automatic Question Answering

With the breakthrough of artificial intelligence in go,"Alpha Go" has entered the public view,and the heat remains high.At the same time,artificial intelligence and deep learning are becoming more and more popular and widely known.One of the major goals of artificial intelligence is to make the machine understand human’s natural language,and it is also the difficult point and focal point at present.Therefore,it is necessary to study the question answering system.And there are many kinds of question answering systems,such as the community question of Baidu Knows and Sogou ask,Knowledge question answering system and Baidu chat robot.These systems,especially knowledge question answering systems,often have knowledge bases.Therefore,it is necessary to study the construction of knowledge base.For high school history question answering system,the significance of the research is to explore the limits of artificial intelligence.The reason for this is that making the machine understand human’s natural language is a difficult point at present.So far,whether the traditional machine learning method is the current hot deep learning,can truly understand human language,and the history of college entrance examination is just ask progress in artificial intelligence detection i n natural language understanding human aspects.The main contents of this paper are as follows:1.First,analyze the history questions of high school,and design the structure of knowledge base according to the characteristics of history questions in high school.2.Construction of historical word segmentation vocabulary,because it is difficult for historical materials to accurately segment,it is often easy to separate related events,professional nouns and so on when the word segmentation,so we need to build a historical word segmentation vocabulary.3.Expansion of high school history knowledge base.Due to the current knowledge base is not informative enough,it is composed of the history textbook,so in order to deal with the history of college entrance examin ation subjects,it is necessary to be extended.4.Content classification of high school history knowledge base.Through the analysis of the history questions of the college entrance examination,the attributes of each historical entity can be roughly divided into seven categories,such as influence category,background category.Therefore,when the knowledge base is expanded,his content should be classified.Use well trained word vectors as input to the classification model,the classification model selects four models such as naive Bias,LSTM,BLSTM,C-LSTM to compare with each other.Finally,the best classifier is selected to classify the content of the high school history knowledge base.