Recommender system is an effective way to help users obtain the personalized and useful information. However, due complexity dynamic, traditional recommender cannot work well in mobile environment. In this paper, we propose a restaurant This adopts user preference model by using features of user's visited restaurants, also utilizes location information restaurants dynamically generate recommendation results. Baidu map cloud service used implement proposed system. The result case study shows...

In this paper we propose a new approach based on text mining techniques for predicting student performance using LSA (latent semantic analysis) and K-means clustering methods. The present study uses free-style comments written by students after each lesson. Since the potentials of these can reflect learning attitudes, understanding subjects difficulties lessons, they enable teachers to grasp tendencies activities. To improve our basic k-means, overlap similarity measuring methods are...

Collaborative filtering (CF) has been widely used in recommender systems to generate personalized recommendations. However, using CF are vulnerable shilling attacks, which attackers inject fake profiles manipulate recommendation results. Thus, attacks pose a threat the credibility of systems. Previous studies mainly derive features from characteristics item ratings user detect attackers, but methods suffer low accuracy when adopt new rating patterns. To overcome this drawback, we properties...

Literature review requires understanding the contents from several view points, such as problem and method that articles describe. Search these viewpoints will improve efficiency of survey, if particular segments were extracted, indexed can be used auxiliary query. This paper focuses on sentences describe in an abstract feature sets classify sentences. Classification performance are evaluated by 10-fold cross-validation for six candidate words. It turned out set all words gains best 90% data...

The bug reports expressed in natural language text usually suffer from vast, ambiguous and poorly written, which causes the challenge to duplicate detection. Current automatic detection techniques have mainly focused on textual information ignored some useful factors. To improve accuracy, this paper, we propose a new approach calls LNG (LDA N-gram) model takes advantages of topic LDA word-based N-gram. considers multiple factors, including information, semantic correlation, word order,...

Search results generated by searchable databases are served dynamically and far larger than the static documents on Web. These pages have been referred to as Deep We need extract target data in integrate them different databases. propose a test bed for information extraction from search results. chose 100 randomly 114,540 with forms. Therefore, these good variety. selected 51 which include URLs pageand manually identify be extracted. also suggest evaluation measures comparing methods extending data.

Analyses on students’ learning behaviors comprise an important thrust in education research. This study focused e-books system used the classroom and this recorded logs their daily academic life. These can be to analysis behaviors. By performing partial correlation analysis, found that a number of have significant relation with test scores.

Predicting students' academic achievement with high accuracy has an important vital role in many disciplines. Most recent studies indicate the of data type selection. They also attempt to understand individual students more deeply by analyzing questionnaire for a particular purpose. The present study uses free-style comments written after each lesson, predict their performance. These reflect learning attitudes understanding subjects, difficulties learn, and activities classroom. To reveal...

To grasp a student's lesson attitude and learning situation to give feed back each student are educational foundations. Goda et al. proposed the PCN method estimate from comment freely written by students[6, 7]. The categorizes comments into three items of P (previous), C(current) N(next).

With the rapid development of information and Web technologies, people are facing 'information overload' in their daily lives.The personalized recommendation system (PRS) is an effective tool to assist users extract meaningful from big data.Collaborative filtering (CF) one most widely used techniques recommend products for users.However, conventional CF technique has some limitations, such as low accuracy similarity calculation, cold start problem, etc.In this paper, a PRS model based on...

We propose a method that constructs hierarchy of words from set given documents automatically and dynamically. Given query, the system retrieves satisfy related are extracted automatically, according their document frequencies. Then hypernym relations these obtained using co-occurrence Empirical evaluation conducted for word derived an English-Japanese dictionary, where descriptions is considered as one documents. It shown meaningful structures obtained. also proposed generates more fine...

The typed λμ-calculus is known to be strongly normalizing and weakly Church-Rosser, hence becomes confluent. In fact, Parigot formulated a parallel reduction prove confluence of the by "Tait-and-Martin-Löf" method. However, diamond property does not hold for his reduction. type-free cannot derived from that confirmed yet as far we know. We analyze granularity rules, then introduce new such both renaming consecutive structural reductions are considered one step It shown formulation has...

Predicting student final score from student's attributes is an important issue of learning analytic. Not only to achieve high prediction performance but also identifying the key research theme. This paper evaluated exhaustively based on all possible combinations four types - behavioral features, demographic academic background, and parent participation. The features are given as numerical data. But, we represented them pair attribute name value. vectorization yields 417 dimensional data,...

This paper compares the prediction performance of document classification based on a variety feature selection measures. Empirical experiments were conducted for dataset re0 with 10 measures and SVM. It is confirmed that SVM-score proposed by Sakai Hirokawa (2012) outperforms standard small number features. In fact, 100 words are enough to get similar obtained all words. The reason good this capture not only characteristic positive samples but negative as well.

Government 2.0 activities have become very attractive and popular these days. Using platforms to support the activities, anyone can anytime report issues or complaints in a city with their photographs geographical information on Web, share them other people. Since variety of reports are posted, officials management section check importance each sort out priorities reports. However, it is not easy task judge When several work task, agreement rate judgments always high. Even if done by only...

In order to overcome mistakes, learners need feedback prompt reflection on their errors.This is a particularly important issue in education systems as the system effectiveness finding errors or mistakes could have an impact learning.Finding essential providing appropriate guidance for flaws.Traditionally task of writing takes time and effort.The authors this paper long-term research goal creating tools learners, especially autonomous enable them be more aware provide way reflect errors.As...

The present study proposes prediction approaches of student's grade based on their comments data. Students describe learning attitudes, tendencies and behaviors by writing freely after each lesson. main difficulty this research is to predict students' performance separately using two class data in Although students learn the same subject, there exist differences between classes. proposed methods basically employ latent semantic analysis (LSA) types machine technique: SVM (support vector...