Event tracing method aiming at news website

Abstract

The invention discloses an event tracing method aiming at a news website. The event tracing method comprises the following steps of: using an event seed report set and an event unrelated report set to train a group of SVM (Support Vector Machine) binary classifiers as an event tracing model; using each SVM binary classifier to classify VSM (Vector Space Model) vectors of main information of a non-processed target news website captured in the news website, so as to obtain a corresponding classifying result; carrying out event related judgment on the target news website according to the classifying result; if the target news website and an event are related, adding the event seed report set, and re-training the event tracing model; and otherwise, continuously processing the new target news website. With the adoption of the event tracing method provided by the invention, the defects of the event tracing method in the prior art of less event related information in an initial stage, data inclination and high computation complexity can be overcome; and the accuracy and the recall rate of the event tracing can be effectively improved.

Description

technical field

[0001] The invention belongs to the technical field of computer data mining, and relates to an event tracking method based on a classification algorithm. Background technique

[0002] In the current situation of network information explosion, due to the rapid update of information and the disorder, it is becoming more and more difficult to find interesting and valuable hot information from the network in time. For this reason, event tracking technology that takes events as the research object has aroused people's interest. Through event tracking, all aspects of event-related information can be automatically organized, thus providing a convenient and quick channel for people to fully understand popular events.

[0003] The current mainstream methods of event tracking methods are divided into the following categories:

[0004] (1) The event tracking model based on the KNN classification algorithm, which first selects the k priori reports that are most similar...

Examples

Embodiment Construction

[0028] The present invention is based on following theoretical basis:

[0029] 1) The event tracking task is based on text classification. This task is usually given 1-4 event seed reports and a set of event unrelated reports. Using these reports, at the event granularity, a classifier is trained for event tracking.

[0030] 2) The more sufficient the prior knowledge, the better the classification effect. The event usually only has 1-4 event seed reports at first, and it is difficult to train a good classification model due to the scarcity of information. Therefore, the present invention uses a search engine to search for event-related information and expand it to the event seed report set, and then utilizes The expanded event seed report set trains the classification model, which can effectively overcome the inaccuracy of the classification model caused by the scarcity of information.

[0031] 3) For statistical classifiers, the classification results will be biased toward c...

technical field

[0001] The invention belongs to the technical field of computer data mining, and relates to an event tracking method based on a classification algorithm. Background technique

[0002] In the current situation of network information explosion, due to the rapid update of information and the disorder, it is becoming more and more difficult to find interesting and valuable hot information from the network in time. For this reason, event tracking technology that takes events as the research object has aroused people's interest. Through event tracking, all aspects of event-related information can be automatically organized, thus providing a convenient and quick channel for people to fully understand popular events.

[0003] The current mainstream methods of event tracking methods are divided into the following categories:

[0004] (1) The event tracking model based on the KNN classification algorithm, which first selects the k priori reports that are most similar...

Examples

Embodiment Construction

[0028] The present invention is based on following theoretical basis:

[0029] 1) The event tracking task is based on text classification. This task is usually given 1-4 event seed reports and a set of event unrelated reports. Using these reports, at the event granularity, a classifier is trained for event tracking.

[0030] 2) The more sufficient the prior knowledge, the better the classification effect. The event usually only has 1-4 event seed reports at first, and it is difficult to train a good classification model due to the scarcity of information. Therefore, the present invention uses a search engine to search for event-related information and expand it to the event seed report set, and then utilizes The expanded event seed report set trains the classification model, which can effectively overcome the inaccuracy of the classification model caused by the scarcity of information.

[0031] 3) For statistical classifiers, the classification results will be biased toward c...

Images