Using of n-grams from morphological tags for fake news classification

Abstract

Research of the techniques for effective fake news detection has become very needed and attractive. These techniques have a background in many research disciplines, including morphological analysis. Several researchers stated that simple content-related n-grams and POS tagging had been proven insufficient for fake news classification. However, they did not realise any empirical research results, which could confirm these statements experimentally in the last decade. Considering this contradiction, the main aim of the paper is to experimentally evaluate the potential of the common use of n-grams and POS tags for the correct classification of fake and true news. The dataset of published fake or real news about the current Covid-19 pandemic was pre-processed using morphological analysis. As a result, n-grams of POS tags were prepared and further analysed. Three techniques based on POS tags were proposed and applied to different groups of n-grams in the pre-processing phase of fake news detection. The n-gram size was examined as the first. Subsequently, the most suitable depth of the decision trees for sufficient generalization was scoped. Finally, the performance measures of models based on the proposed techniques were compared with the standardised reference TF-IDF technique. The performance measures of the model like accuracy, precision, recall and f1-score are considered, together with the 10-fold cross-validation technique. Simultaneously, the question, whether the TF-IDF technique can be improved using POS tags was researched in detail. The results showed that the newly proposed techniques are comparable with the traditional TF-IDF technique. At the same time, it can be stated that the morphological analysis can improve the baseline TF-IDF technique. As a result, the performance measures of the model, precision for fake news and recall for real news, were statistically significantly improved.

The Use of Syntactic Information in fake news detection: A Systematic Review

“…The sophisticated architecture with syntactic features proposed by Gupta et al [2021] presented an increase of up to 3 % in the performance. Kapusta et al [2021] noticed an accuracy increase by 3 to 4 %, where the average accuracies for syntactic, readability, and combined features are 84.12%, 77.67%, and 84.52%, respectively. Nguyen et al [2019] identified that n-grams in combination with dependency sub-trees as features have a positive impact on the performance of the classifier.…”

“…Zhou et al [2020] saw no improvement with shallow syntax, but deep syntax-level features (CFGs) and features at lexicon-level (BOWs) outperform the others. Kapusta et al [2021] concluded that morphological analysis can be applied to fake news classification.…”

Using a classification model for determining the value of liver radiological reports of patients with colorectal cancer

“…The occurrence frequency of all grams is counted and filtered according to the preset threshold to form a list of key grams. N-gram language model shows good performance in many text mining tasks (30)(31)(32). For example, Giannakopoulos and Karkaletsis (30) expressed the text as an n-gram model using a sliding window with a length of n by connecting the adjacent n-grams with the edges representing their co-occurrence frequency in a given text window, they captured the word order in the text and detected some similarities in the text morphology.…”

A review on authorship attribution in text mining

“…This accuracy was lower than that obtained using word n ‐grams (96.8%) and higher than that obtained using character n ‐grams (87.1%). Zafarani et al (2019) and Kapusta et al (2021) used parts of POS tag n ‐grams as morphological characteristics of words to detect fake news. POS tag n ‐grams can also be used to predict author personality (Litvinova et al, 2015).…”