The Detection of Fake Reviews in Bestselling Books

This study detected the possible manipulation of reviews for bestseller books. The authors first used clustering analysis to identify the cluster of bestselling books and patterns of manipulated reviews and ratings. They then used an artificial neural network to predict the possibility of review manipulation in bestselling books based on the patterns identified. The prediction outcome has an accuracy rate of 89%. They found that fake or manipulated reviews for bestselling books could be identified by analyzing abnormal rating fluctuations. The findings could help e-commerce platforms identify review manipulations and thereby help customers make prudent purchase decisions.

223 The Clinical Applications of Machine Learning in Plastic and Reconstructive Surgery: A Systematic Review

Background Machine learning (ML) is a set of models and methods that can automatically detect patterns in vast amounts of data and use this information to perform various kinds of decision-making under uncertain conditions. The aim of this review is to evaluate the applications of machine learning in plastic and reconstructive surgery. Method EMBASE, MEDLINE and CENTRAL were searched from 1990 to 2020 for studies in which machine learning has been employed in the clinical setting of reconstructive surgery. Primary outcomes will be the evaluation of the accuracy of machine learning models in predicting a clinical diagnosis and post-surgical outcomes. Results The database identified 1181 articles, of which 51 articles were included in this review. The clinical utility of these algorithms was to assist clinicians in diagnosis prediction (n = 22), outcome prediction (n = 21) and pre-operative planning (n = 8). The mean accuracy for diagnosis prediction, outcome prediction and pre-operative planning was 88.80%, 86.11% and 80.28% respectively. The most commonly used models were neural networks (n = 31), support vector machine (n = 13), decision trees/random forests (10) and logistic regression (n = 9). Discussion ML has demonstrated excellent performance in diagnosis and outcome predictions, but it is still in its infancy. Further research is warranted to evaluate its applications.

Literature Mining for Incorporating Inductive Bias in Biomedical Prediction Tasks (Student Abstract)

Applications of machine learning in biomedical prediction tasks are often limited by datasets that are unrepresentative of the sampling population. In these situations, we can no longer rely only on the the training data to learn the relations between features and the prediction outcome. Our method proposes to learn an inductive bias that indicates the relevance of each feature to outcomes through literature mining in PubMed, a centralized source of biomedical documents. The inductive bias acts as a source of prior knowledge from experts, which we leverage by imposing an extra penalty for model weights that differ from this inductive bias. We empirically evaluate our method on a medical prediction task and highlight the importance of incorporating expert knowledge that can capture relations not present in the training data.

Prediction Outcome for Massive Multiplayer Online Games Using Data Mining

Massive Multiplayer Online (MMO) game is one of the famous game genres among teenagers nowadays. MMO games allow gamers to interact and play with up to thousand players. Rainbow Six Siege (RSS) belongs to MMO type of game. However, due to many operators that are available in this game, the player needs to choose the right operator to counter the enemy operator. Therefore, based on the characteristic of the selected operator, this paper attempted to predict the outcomes of the game. In our prediction model, characteristics for these operators were extracted from 120 live stream replays. Three classification algorithms were utilized to predict the outcome of the game. Among these algorithms, IBK had obtained outstanding performance in the dataset. The accuracy of the model is 93.75%, applying 5-fold cross-validation test. The success rate reveals that our proposed model is suitable to predict the outcome of the game.