Search

Pseudo-relevance feedback (PRF) is a relevance feedback (RF) technique for information retrieval that treats the top k retrieved images as relevance feedback. PRF is used to avoid the limitations o...

Digital dictionaries have been widely used to facilitate word processing and word search through digital media such as mobile phones. Society generally knows the efficacy and how to mix medicinal plants from the experience of previous parents or through books and writings. Searching through books or writings requires a short time compared to searching through digital media, one of which is a digital dictionary. The research aims to create a digital dictionary of mobile-based medicinal plants which has a search facility based on the words entered, for example, the contents of the medicinal plants. The digital dictionary application of medicinal plants uses a search method with the Rocchio algorithm with relevance feedback techniques to check the proximity of the query to the average document relevant to the level of similarity calculation through the stages of tokenizing, filtering, stemming, and Term Weighting with a total data of 200 medicinal plants.

Objective With increasing patient complexity whose data are stored in fragmented health information systems, automated and time-efficient ways of gathering important information from the patients' medical history are needed for effective clinical decision making. Using COVID-19 as a case study, we developed a query-bot information retrieval system with user-feedback to allow clinicians to ask natural questions to retrieve data from patient notes. Materials and methods We applied clinicalBERT, a pre-trained contextual language model, to our dataset of patient notes to obtain sentence embeddings, using K-Means to reduce computation time for real-time interaction. Rocchio algorithm was then employed to incorporate user-feedback and improve retrieval performance. Results In an iterative feedback loop experiment, MAP for final iteration was 0.93/0.94 as compared to initial MAP of 0.66/0.52 for generic and 1./1. compared to 0.79/0.83 for COVID-19 specific queries confirming that contextual model handles the ambiguity in natural language queries and feedback helps to improve retrieval performance. User-in-loop experiment also outperformed the automated pseudo relevance feedback method. Moreover, the null hypothesis which assumes identical precision between initial retrieval and relevance feedback was rejected with high statistical significance (p ≪ 0.05). Compared to Word2Vec, TF-IDF and bioBERT models, clinicalBERT works optimally considering the balance between response precision and user-feedback. Discussion Our model works well for generic as well as COVID-19 specific queries. However, some generic queries are not answered as well as others because clustering reduces query performance and vague relations between queries and sentences are considered non-relevant. We also tested our model for queries with the same meaning but different expressions and demonstrated that these query variations yielded similar performance after incorporation of user-feedback. Conclusion In conclusion, we develop an NLP-based query-bot that handles synonyms and natural language ambiguity in order to retrieve relevant information from the patient chart. User-feedback is critical to improve model performance.

Current search engines like Google, Yahoo give the search results, indeed users are facing problems in information retrieval. The main problem is because of word mismatch and availability of many resources. Petabytes of information is available because of Internet. From that huge available data, for a naive user it becomes hectic to distinguish between relevant and irrelevant information to individual interest. Also, another reason of getting irrelevant information is incompatibility between terms that users are using and keywords present in documents. Query expansion is an adding keyword to the original query. The main issue in query expansion is selection of appropriate terms from user’s original query. Vocabulary database helps to solve this issue. Identification of the similar words and language entities that are similar in meaning is done by vocabulary which is frequently incorporated in information retrieval system. Thesaurus has been used across a large area of in information retrieval also many applications and natural language processing. In this work, to improve performance of a search query, BM25 model is used for query expansion. Cosine similarity is used to determine similarity between two keywords. Rocchio algorithm is used to calculate the relevance feedback. Experimental result shows better results using Rocchio algorithm.

When searching for information with an information retrieval (IR) system, sometimes the results of the search documents provided by the system do not match the information needs of the user. Pseudo Relevance Feedback (PRF) based Query expansion (QE) tries to overcome these problems by adding words that are expected to improve retrieval results from top N ranked documents retrieved. The use of firefly algorithm (FA) as one of the optimization methods has been proven by the previous study to improve the performance of the IR system. However, in that study the weighting of words was done using the rocchio function of the Pseudo Relevant Document (PRD), so it is feared that the performance of IR system will be reduced if the number of relevant documents in PRD is little or none at all. Therefore, scoring by term relationship between query and PRD is used in this study combined with rocchio algorithm. The results of the study showed that usage of term relationship word co-occurrence or word similarity can improve the performance of the IRS that was previously built. In addition, word co-occurrence with jaccard have the best performance compared to the previous study and other combinations. FA itself was able to choose the optimal terms, even though the number of top N ranked documents increased. Furthermore, the combination of term relationship and rocchio algorithm can increase the convergence rate than the ones without rocchio algorithm.

Pseudo relevance feedback (PRF) enhances the retrieval performance of the relevance feedback. Pseudo relevance feedback assumes that the k highest-ranking documents in the first retrieval are relevant and extract query expansion from them. Rocchio algorithm is a classical algorithm for implementing relevance feedback into vector space models. The Rocchio algorithm forms a new query moves toward the centroid of the relevant documents and keeps away from centroid of the irrelevant documents. However, in the relevance feedback method, irrelevant documents are ignored. In this paper, we conduct a method for pseudo irrelevance feedback (PIRF) documents components that effectively applied to the Rocchio algorithm. Documents with a high ranking outside of k relevant documents and those documents dissimilar to any k relevant documents can extract good query expansion if the documents are applied as irrelevant documents. The Rocchio algorithm uses PRF as a component of relevant documents and this research method for irrelevant documents as a component of irrelevant documents denoted by Roc PRF PIRF (filter). Experiment on CISI dataset show that Roc PRF PIRF (filter) improved performance by testing several variations the number of irrelevant documents compared to the standard Rocchio algorithm and Rocchio algorithm with irrelevant documents but without proposed method).

This paper explores the effectiveness of Particle Swarm Classification (PSC) for a classification task in the field of educational data mining. More specifically, it proposes PSC to design a classification model capable of classifying questions into the six cognitive levels of Bloom's taxonomy. To this end, this paper proposes a novel specialized initialization mechanism based on Rocchio Algorithm (RA) to mitigate the adverse effects of the curse of dimensionality on the PSC performance. Furthermore, in the design of the RA-based PSC model of questions classification, several feature selection approaches are investigated. In doing so, a dataset of teachers' classroom questions was collected, annotated manually with Bloom's cognitive levels, and transformed into a vector space representation. Using this dataset, several experiments are conducted, and the results show a poor performance of the standard PSC due to the curse of dimensionality. However, when the proposed RA-based initialization mechanism is used, a significant improvement in the average performance, from 0.243 to 0.663, is obtained. In addition, the results indicate that the feature selection approaches play a role in the performance of the RA-based PSC (average performance ranges from 0.535 to 0.708). Finally, a comparison between the performance of RA-based PSC (average performance = 0.663) and seven machine learning approaches (best average performance = 0.646) confirms the effectiveness of the proposed RA-based PSC approach. Keywords: Particle swarm classification, Rocchio Algorithm, Educational data mining, Questions classification, Bloom's taxonomy

To support complex search tasks, where the initial information requirements are complex or may change during the search, a search engine must adapt the information delivery as the user's information requirements evolve. To support this dynamic ranking paradigm effectively, search result ranking must incorporate both the user feedback received, and the information displayed so far. To address this problem, we introduce a novel reinforcement learning-based approach, RLIrank. We first build an adapted reinforcement learning framework to integrate the key components of the dynamic search. Then, we implement a new Learning to Rank (LTR) model for each iteration of the dynamic search, using a recurrent Long Short Term Memory neural network (LSTM), which estimates the gain for each next result, learning from each previously ranked document. To incorporate the user's feedback, we develop a word-embedding variation of the classic Rocchio Algorithm, to help guide the ranking towards the high-value documents. Those innovations enable RLIrank to outperform the previously reported methods from the TREC Dynamic Domain Tracks 2017 and exceed all the methods in 2016 TREC Dynamic Domain after multiple search iterations, advancing the state of the art for dynamic search., Proceedings of The Web Conference 2020 (WWW '20), April 20--24, 2020, Taipei, Taiwan

This paper proposes a new variant of Particle Swarm Optimization (PSO), dubbed CentroidPSO, to tackle data classification problem in high dimensional domains. It is inspired by the center-based sam...

In recent years, there has been a growing interest in applying Particle Swarm Optimization (PSO) to data classification. Nonetheless, due to the curse of dimensionality, the effectiveness of the PSO applied to high dimensional data classification becomes questionable. This paper proposes a novel specialized PSO initialization mechanism, developed specifically for PSO applications to high dimensional data classification. The proposed initialization mechanism is inspired by the center-based sampling theory, which argues that the center of the search space is a promising region for the initialization step in evolutionary algorithms. Furthermore, the proposed initialization mechanism is based on an information retrieval algorithm called Rocchio Algorithm (RA); that identifies the center region of the search space of data classification. To validate the proposed mechanism, RA-based PSO has been applied to a high dimensional classification task in educational data mining. More specifically, RA-based PSO has been applied to classify a dataset of teachers' classroom questions into Bloom's taxonomy cognitive levels. To do so, a dataset of teachers' classroom questions has been collected and annotated manually with Bloom's taxonomy cognitive levels. Pre-processing steps have been applied to convert questions into a representation suitable for classification. Using this dataset, the standard PSO, PSO with generic initialization mechanisms, and RA-based PSO have been experimented and compared. The results show a poor performance of the standard PSO and the PSO with the generic initialization mechanisms, as well as a significant improvement in the performance of RA-based PSO. These results indicate that a proper task-specific PSO initialization mechanism is crucial for effective PSO performance in high dimensional data classification. Furthermore, a comparison between RA-based PSO and pure RA classification provide a quantitative estimation of the role of initialization mechanism and PSO search for the classification of the dataset. On the other hand, the comparison between RA-based PSO approach and three conventional machine learning approaches, experimented on the same dataset confirms the effectiveness of RA-based PSO for high dimensional data classification. Moreover, the comparison between RA-based PSO approach and machine learning approaches, in terms of computational time efficiency, shows that they are comparable in classification time. However, as the learning of PSO is a time-consuming process, its effectiveness is significantly affected if the learning time is a matter.

Accurate, efficient and fast processing of textual data and classification of electronic documents have become an important key factor in knowledge management and related businesses in today’s world. Text mining, information retrieval, and document classification systems have a strong positive impact on digital libraries and electronic content management, e-marketing, electronic archives, customer relationship management, decision support systems, copyright infringement, and plagiarism detection, which strictly affect economics, businesses, and organizations. In this study, we propose a new similarity measure that can be used with k-nearest neighbors (k-NN) and Rocchio algorithms, which are some of the well-known algorithms for document classification, information retrieval, and some other text mining purposes. We have tested our novel similarity measure with some structured textual data sets and we have compared the results with some other standard distance metrics and similarity measures such as Cosine similarity, Euclidean distance, and Pearson correlation coefficient. We have obtained some promising results, which show that this proposed similarity measure could be alternatively used within all suitable algorithms, methods, and models for text mining, document classification, and relevant knowledge management systems. Keywords: text mining, document classification, similarity measures, k-NN, Rocchio algorithm

This paper presents Content based Image Retrieval with Relevance Feedback to retrieve relevant images based on an image query. Three main steps are proposed, first, obtain 2D feature representation of an image query and image database using the Integrated Color Co-Occurrence Matrix. This feature extraction method captures two features simultaneously, they are color and texture features. Second, compute cosine similarity measurement to retrieve similar images between features of an image query and features of all images in the database. Third, update the query features using Rocchio algorithm based on the user's relevance feedback, and recalculation of the cosine similarity between the updated feature of query and features of all images in the database. Experiments are conducted using Corel Image database that consists of 1000 images from ten classes. The proposed model for retrieving similar images achieved higher performance accuracy compare to the Content based Image Retrieval without Relevance feedback.

Books are the source of knowledge and information. Users search relevant books based on their need. In this paper we develop method for book suggestion. In this paper, we combine Sequential Dependence Model with heterogeneous social information such as tags, reviews and ratings. These social informations are assigned by users. Therefore, we use CLEF-2016 Social Book Search track (Suggestion task). We train our proposed method on CLEF-2015 Social Book Search (Suggestions task) datasets and tested on CLEF-2016 Social book Search datasets. We get better result compared to other systems.

The amount of information that is available on internet related to music is very huge. The information related to music can be mined using many features, and the on-line contribution of both musical experts and general listeners has provided music researchers with a rich resource of information. The mood of a song helps in recommending songs to online users. Also, there is a strong application-oriented interest in mood classification for music download services and audio players allow music collection browsing using mood as one search criteria. This paper proposes a novel mood classification technique using improved Roccihio algorithm. Since Rocchio algorithm uses only one prototype vector for representing a class, it offers a less prediction accuracy. We addressed this problem by considering the $k$ -nearest vectors along with prototype vector. The proposed method is validated using real music data, collected from well known music portals, in comparison with other machine learning methods.

Traditional text classification algorithms have vital impact on information filtering. However, their performances were confined to a large extent in terms of the massive data set. This paper proposes an approach using MapReduce-based Rocchio relevance feedback algorithm, which improved the traditional Rocchio algorithm in the MapReduce paradigm, to resolve the problem of massive information filtering. The experiments on Hadoop cluster showed an effective improvement in performance by using the new method.

The need for the documents in the form of journals or scientific articles is currently increasing which means the need of documents available also increasing, in another side it makes difficult to find and present journals. Therefore, a method is needed to classify journals automatically according to the categories of the journal. One method that can help organize documents according to their categories is classification. In this study, Rocchio algorithm is used as a method to classify journals. The data used comes from Google Scholar. Journal consists of 4 (four) categories: Computer Vision and Pattern Recognition, Artificial Intelligent, Data Mining and Analysis, and Computer System. System testing is carried out by taking 160 journals for training data and 48 journals for testing data. The results of this study produced a web-based journal classification system and showed the method used was able to classify journals with accuracy result is 93%.

Millions of file uploads and downloads happen every minute resulting in big data creation and manual text categorization is not possible. Hence, there is a need for automatic categorization of documents that makes storage and retrieval more efficient. This research paper proposes a hybrid text categorization model that combines both Rocchio algorithm and Random Forest algorithm to perform Multi-label text categorization. Stop word remover and word stemmer has been used to overcome the limitations in Rocchio Algorithm. Random Forest model takes minimal categories as input to reduce its error rate. Experiments were done on standard text categorization datasets. Our proposed model is found to be more efficient in categorizing the documents when compared with other text categorization models such as fuzzy relevance clustering, ML-KNN (Multi-label KNN) and Naive-Bayes Algorithms.

Hierarchical data is becoming increasingly prominent, especially on the web. Wikipedia is one such example where there are millions of documents that are classified into multiple classes in a hierarchical fashion. This gives rise to an interesting problem of automating the classification of new documents. As the size of the dataset grows, so does the number of classes. Further, there seems to be sparsity issue even with the increase in the dataset. Therefore, this poses a challenge to classify data in such a manner. We present two different algorithms based on text categorization: Rocchio algorithm and kNN. We implement and compare the above mentioned methods to better understand the approach to take in classifying hierarchical data.

This paper proposes an approach using MapReduce-based Rocchio relevance feedback algorithm, which improved the traditional Rocchio algorithm in the MapReduce paradigm, to resolve the problem of massive information filtering. Traditional text classification algorithms have vital impact on information filtering.

In the modern world, networked sensor technology makes it possible to capture the world around us in real-time. In the security domain cameras are an important source of information. Cameras in public places, bodycams, drones and recordings with smart phones are used for real time monitoring of the environment to prevent crime (monitoring case); and/or for investigation and retrieval of crimes, for example in evidence forensics (forensic case). In both cases it is required to quickly obtain the right information, without having to manually search through the data. Currently, many algorithms are available to index a video with some pre-trained concepts, such as people, objects and actions. These algorithms require a representative and large enough set of examples (training data) to recognize the concept. This training data is, however, not always present. In this thesis, we aim to assist an analyst in their work on video stream data by providing a search capability that handles ad-hoc textual queries, i.e. queries that include concepts or events that are not pre-trained. We use the security domain as inspiration for our work, but the analyst can be working in any application domain that uses video stream data, or even indexed data. Additionally, we do only consider the technical aspects of the search capability and not on the legal, ethical or privacy issues related to video stream data. We focus on the retrieval of high-level events, such as birthday parties. We assume that these events can be composed of smaller pre-trained concepts, such as a group of people, a cake and decorations and relations between those concepts, to capture the essence of that unseen event (decompositionality assumption). Additionally, we hold the open world assumption, i.e. the system does not have complete world knowledge. Although current state of the art systems are able to detect an increasingly large number of concepts, this number still falls far behind the near infinite number of possible (textual) queries that a system needs to be able to handle. In our aim to assist the analyst, we focus on the improvement of the visual search effectiveness (e.g. performance) by a semantic query-to-concept mapping: the mapping from the user query to the set of pre-trained concepts. We use the TRECVID Multimedia Event Detection benchmark, as it contains high-level events inspired by the security domain. In this thesis, we show that the main improvements can be achieved by using a combination of i) queryto- concept mapping based on semantic word embeddings (+12%), ii) exploiting user feedback (+26%) and iii) fusion of different modalities (data sources) (+17%). First, we propose an incremental word2vec (i-w2v) method [1], which uses word2vec trained on GoogleNews items as a semantic embedding model and incrementally adds concepts to the set of selected concepts for a query in order to deal with query drift. This method improve performance in terms of MAP compared to the state of the art word2vec method and knowledge based techniques. In combination with a state of the art video event retrieval pipeline, we achieve top performance on the TRECVID MED benchmark regarding the zero-example task (MED14Test results). This improvement is, however, dependent on the availability of the concepts in the Concept Bank: without concepts related to or occurring in the event, we cannot detect the event. We, thus, need a properly composed Concept Bank to properly index videos. Second, we propose an Adaptive Relevance Feedback interpretation method named ARF [2] that not only achieves high retrieval performance, but is also theoretically founded through the Rocchio algorithm from the text retrieval field. This algorithm is adjusted to the event retrieval domain in a way that the weights for the concepts are changed based on the positive and negative annotations on videos. The ARF method has higher visual search effectiveness compared to k-NN based methods on video level annotations and methods based on concept level annotations. Third, we propose blind late fusion methods that are based on state of the art methods [3], such as average fusion or fusion based on probabilities. Especially the combination of a Joint Ratio (ratio of probabilities) and Extreme Ratio (ratio of minimum and maximum) method (JRER) achieves high performance in cases with reliable detectors, i.e. enough training examples. This method is not only applicable to the video retrieval field, but also in sensor fusion in general. Although future work can be done in the direction of implicit query-to-concept mapping through deep learning methods, smartly combining the concepts and the usage of spatial and temporal information, we have shown that our proposed methods can improve the visual search effectiveness by a semantic query-to-concept mapping which brings us a step closer to a search capability that handles ad-hoc textual queries for analysts.

In the text literature, many topic models were proposed to represent documents and words as topics or latent topics in order to process text effectively and accurately. In this paper, we propose LDACLM or Latent Dirichlet Allocation Category LanguageModel for text categorization and estimate parameters of models by variational inference. As a variant of Latent Dirichlet Allocation Model, LDACLM regards documents of category as Language Model and uses variational parameters to estimate maximum a posteriori of terms. In general, experiments show LDACLM model is effective and outperform Na¨?ve Bayes with Laplace smoothing and Rocchio algorithm but little inferior to SVM for text categorization.

This study examines various weighting methods for improving the performance of automatic classification based on Rocchio algorithm on two collections(LISA, Reuters-21578). First, three factors for weighting are identified as document factor, document factor, category factor for each weighting schemes, the performance of each was investigated. Second, the performance of combined weighting methods between the single schemes were examined. As a result, for the single schemes based on each factor, category-factor-based schemes showed the best performance, document set-factor-based schemes the second, and document-factor-based schemes the worst. For the combined weighting schemes, the schemes(idf*cat) which combine document set factor with category factor show better performance than the combined schemes(tf*cat or ltf*cat) which combine document factor with category factor as well as the common schemes (tfidf or ltfidf) that combining document factor with document set factor. However, according to the results of comparing the single weighting schemes with combined weighting schemes in the view of the collections, while category-factor-based schemes(cat only) perform best on LISA, the combined schemes(idf*cat) which combine document set factor with category factor showed best performance on the Reuters-21578. Therefore for the practical application of the weighting methods, it needs careful consideration of the categories in a collection for automatic classification.

International audience; Recommender system provides relevant items to users from huge catalogue. Collaborative filtering and content-based filtering are the most widely used techniques in personalized recommender systems. Collaborative filtering uses only the user-ratings data to make predictions, while content-based filtering relies on semantic information of items for recommendation. Hybrid recommendation system combines the two techniques. In this paper, we present another hybridization approach: User Semantic Collaborative Filtering. The aim of our approach is to predict users preferences for items based on their inferred preferences for semantic information of items. In this aim, we design a new user semantic model to describe the user preferences by using Rocchio algorithm. Due to the high dimension of item content, we apply a latent semantic analysis to reduce the dimension of data. User semantic model is then used in a user-based collaborative filtering to compute prediction ratings and to provide recommendations. Applying our approach to real data set, the MoviesLens 1M data set, significant improvement can be noticed compared to usage only approach, content based only approach.

Exploring digital collections to find information relevant to a user's interests is a challenging task. Algorithms designed to solve this relevant information problem base their relevance computations on user profiles in which representations of the users' interests are maintained. This article presents a new method, based on the classic Rocchio algorithm for text categorization, able to discover user preferences from the analysis of textual descriptions of items in online catalog of e-commerce Web sites. Experiments have been carried out on several data sets, and results have been compared with those obtained using an inductive logic programming (ILP) approach and a probabilistic one.

Several performance factors which have applied to the automatic indexing with controlled vocabulary and text categorization based on Rocchio algorithm were examined, and the simple method for performance improvement of them were tried. Also, results of the methods using Rocchio algorithm were compared with those of other learning based methods on the same conditions. As a result, keeping with the strong points which are implementational easiness and computational efficiency, the methods based Rocchio algorithms showed equivalent or better results than other learning based methods(SVM, VPT, NB). Especially, for the semi-automatic indexing(computer-aided indexing), the methods using Rocchio algorithm with a high recall level could be used preferentially.

Information filtering (IF) systems usually filter data items by correlating a set of terms representing the user's interest (a user profile) with similar sets of terms representing the data items. Many techniques can be employed for constructing user profiles automatically, but they usually yield large sets of term. Various dimensionality-reduction techniques can be applied in order to reduce the number of terms in a user profile. We describe a new terms selection technique including a dimensionality-reduction mechanism which is based on the analysis of a trained artificial neural network (ANN) model. Its novel feature is the identification of an optimal set of terms that can classify correctly data items that are relevant to a user. The proposed technique was compared with the classical Rocchio algorithm. We found that when using all the distinct terms in the training set to train an ANN, the Rocchio algorithm outperforms the ANN based filtering system, but after applying the new dimensionality-reduction technique, leaving only an optimal set of terms, the improved ANN technique outperformed both the original ANN and the Rocchio algorithm.

Keeping track of changing interests is a natural phenomenon as well as an interesting tracking problem because interests can emerge and diminish at different time frames. Being able to do so with a few feedback examples poses an even more important and challenging problem because existing concept drift learning algorithms that handle the task typically suffer from it. This work presents a new computational framework for extending incomplete labeled data stream (FEILDS), which extends the capability of existing algorithms for learning concept drift from a few labeled data. The system transforms the original input stream into a new stream that can be conveniently tracked by the existing learning algorithms. The experiment results reveal that FEILDS can significantly improve the performances of a Multiple Three-Descriptor Representation (MTDR) algorithm, Rocchio algorithm, and window-based concept drift learning algorithms when learning from a sparsely labeled data stream with respect to their performances without using FEILDS.

The goal of this paper is to derive extra representatives from each class to compensate for the potential weakness of linear classifiers that compute one representative for each class. To evaluate the effectiveness of our approach, we compared with linear classifier produced by Rocchio algorithm and the k-nearest neighbor (kNN) classifier. Experimental results show that our approach improved linear classifier and achieved micro-averaged accuracy close to that of kNN, with much less classification time. Furthermore, we could provide a suggestion to reorganize the structure of classes when identify new representatives for linear classifier.

This paper presents the design and evaluation of a text categorization method based on the Hierarchical Mixture of Experts model. This model uses a divide and conquer principle to define smaller categorization problems based on a predefined hierarchical structure. The final classifier is a hierarchical array of neural networks. The method is evaluated using the UMLS Metathesaurus as the underlying hierarchical structure, and the OHSUMED test set of MEDLINE records. Comparisons with an optimized version of the traditional Rocchio's algorithm adapted for text categorization, as well as flat neural network classifiers are provided. The results show that the use of the hierarchical structure improves text categorization performance with respect to an equivalent flat model. The optimized Rocchio algorithm achieves a performance comparable with that of the hierarchical neural networks.

International audience; Recommender system provides relevant items to users from huge catalogue. Collaborative filter-ing and content-based filtering are the most widely used techniques in personalized recommender systems. Collaborative filtering uses only the user-ratings data to make predictions, while content-based filtering relies on semantic information of items for recommendation. Hybrid recommenda-tion system combines the two techniques. The aim of this work is to introduce a new approach for semantically enhanced collaborative filtering. Many works have addressed this problem by proposing hybrid solutions. In this paper, we present another hybridization technique that pre-dicts users preferences for items based on their inferred preferences for semantic information of items. For this, we design a new user semantic model by using Rocchio algorithm and we apply a latent semantic analysis to reduce the dimension of data. Applying our approach to real data, the MoviesLens 1M dataset, significant improvement can be noticed compared to usage only approach, and hybrid algorithm.

Sentiment analysis is an important issue in machine learning, which aims to identify the emotion expressed in corpus. However, sentiment analysis is a difficult task, especially in large-scale data, where feature reduction is needed. In this paper, we propose a parallel feature reduction algorithm for sentiment polarity classification based on a substring method. Specifically, the proposed algorithm is based on parallel computing under the Hadoop platform. The proposed algorithm is examined on a large data set and a K-nearest neighbor algorithm and a Rocchio algorithm are used for classification. Experimental results show that the proposed algorithm outperforms other commonly used methods in terms of the classification performance and the computational cost.

Text categorization is used to assign each text document to predefined categories. This paper presents a new text classification method for classifying Chinese text based on Rocchio algorithm. We firstly use the TFIDF to extract document vectors from the training documents which have been correctly categorized, and then use those document vectors to generate codebooks as classification models using the LBG and Rocchio algorithm. The codebook is then used to categorize the target documents using vector scores. We tested this method in the experiment and the result shows that this method can achieve better performance.