Transformation of discriminative single-task classification into generative multi-task classification in machine learning context

Abstract

Classification is one of the most popular tasks of machine learning, which has been involved in broad applications in practice, such as decision making, sentiment analysis and pattern recognition. It involves the assignment of a class/label to an instance and is based on the assumption that each instance can only belong to one class. This assumption does not hold, especially for indexing problems (when an item, such as a movie, can belong to more than one category) or for complex items that reflect more than one aspect, e.g. a product review outlining advantages and disadvantages may be at the same time positive and negative. To address this problem, multi-label classification has been increasingly used in recent years, by transforming the data to allow an instance to have more than one label; the nature of learning, however, is the same as traditional learning, i.e. learning to discriminate one class from other classes and the output of a classifier is still single (although the output may contain a set of labels). In this paper we propose a fundamentally different type of classification in which the membership of an instance to all classes(/labels) is judged by a multiple-inputmultiple-output classifier through generative multi-task learning. An experimental study is conducted on five UCI data sets to show empirically that an instance can belong to more than one class, by using the theory of fuzzy logic and checking the extent to which an instance belongs to each single class, i.e. the fuzzy membership degree. The paper positions new research directions on multitask classification in the context of both supervised learning and semi-supervised learning.

Machine Classification for Suicide Ideation Detection on Twitter

“…Research related to suicide-related communications on social media are still in primary level; For instance, [3][4] [5] noticed the association of suicide ideation , social networks, followed by elevating consequences related to the safety of people and creating more platforms for doing further analysis to serve users who are in danger. By using the machine learning classifiers, the authors generated a list of words related to suicide ideation in order to perform the classification of suicidal and non-suicidal texts , and acquired an accuracy of 60%.…”

“…Even though these sites are beneficial, they are creating negative impact on people having suicidal thoughts [2]. Several researches reported the relationship between group of people with suicidal ideation and social networks [3][4] [5]. Now-a-days people are killing themselves based on the text received in social media.…”

“…It has been used in different areas like designing of drugs, detection of frauds and searching in web and online recommender systems. Along with this, the most effective technique of ML is classification [3][4] [5], which performs the judgment of new attribute.…”

Fuzzy Multi-task Learning for Hate Speech Type Identification

“…In general, different labels can have specific relationships such as mutual independence and positive correlation (co-occurrence) , if the labels are not mutually exclusive. In this context, if the labels assigned to each instance are mostly independent of each other, the above Label Power-set strategy for handling multi-labelled data could result in an exponential increase of the number of classes and a reduced number of instances for each class [22,32]. In this case, the complexity of the learning task is much increased leading to a higher risk of overfitting [29].…”

Suiciderelated Text Classification With Prism Algorithm

“…According to [3], the use of machine learning has spread rapidly in the last decade especially in computer science, as it has been applied to various and diverse areas such as fraud detection, drug design, web search and recommender systems. Furthermore, one of the most popular tasks in machine learning is classification [3][4][5], where the category of an unseen instance is judged.…”