To grasp a student's lesson attitude and learning situation and to give a feed back to each student are educational foundations. Goda et al. proposed the PCN method to estimate a learning situation from a comment freely written by students [6,7]. The PCN method categorizes comments into three items of P(previous), C (current) and N(next).They pointed out a correlation between the student's final results and the validity of a descriptive content of item C, that is something related to understanding of the lesson and learning attitudes to the lesson. However, a problem left in their work is the badness of performance in prediction for upper grade students.This paper proposes two manners of utilization of PCN scores: the validity level determination for assessment, and for prediction performance of students' final grades.In order to validate the proposed manners of utilization, we conducted two experiments. First, we employed multiple regression analysis to calculate PCN scores that determine the validity level with respect to each viewpoint. Students who wrote comments with a high PCN score are considered as those who describe their learning attitude appropriately. We also applied a machine learning method SVM (support vector machine) to students' comments for predicting their final results in five grades of S, A, B, C and D.Experimental results illustrated that as comments of students get higher PCN scores, the prediction performance of the students' grades becomes higher.
SUMMARYCollaborative filtering (CF) has been widely used in recommender systems to generate personalized recommendations. However, recommender systems using CF are vulnerable to shilling attacks, in which attackers inject fake profiles to manipulate recommendation results. Thus, shilling attacks pose a threat to the credibility of recommender systems. Previous studies mainly derive features from characteristics of item ratings in user profiles to detect attackers, but the methods suffer from low accuracy when attackers adopt new rating patterns. To overcome this drawback, we derive features from properties of item popularity in user profiles, which are determined by users' different selecting patterns. This feature extraction method is based on the prior knowledge that attackers select items to rate with man-made rules while normal users do this according to their inner preferences. Then, machine learning classification approaches are exploited to make use of these features to detect and remove attackers. Experiment results on the MovieLens dataset and Amazon review dataset show that our proposed method improves detection performance. In addition, the results justify the practical value of features derived from selecting patterns.
We present grammatical (or equational) descriptions of the set of normal inhabitants fM j 0`M : A; M in-normal form g of a given type A under a given basis 0, both for the standard simple type system (in partial discharge convention) and for the system in total discharge convention (or Prawitz-style natural deduction system). It is shown that in the latter system we can describe the set by a (nite) context-free grammar, but for the standard system it is not necessarily the case because we may need an innite supply of fresh (bound) variables to describe the set. In both cases, however, our grammars reect the structure of normal inhabitants in such a way that, when non-terminals are ignored, a derivation tree of the grammars yielding a-term M can be identied with the B ohm tree of M. We give some applications of the grammatical descriptions. Among others, we give simple algorithms for the emptyness/niteness problem of the set of normal inhabitants of a given type (both for the standard and nonstandard systems).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.