Figure 2 - uploaded by Miao Fan
Content may be subject to copyright.
Source publication
Content-based social tagging recommendation, which considers the relationship between the tags and the descriptions contained in resources, is proposed to remedy the cold-start problem of collaborative filtering. There is such a common phenomenon that certain tag does not appear in the corresponding description, however, they do semantically relate...
Context in source publication
Similar publications
Providing useful information to the users by recommending highly demanded products and services is a fundamental part of the business of many top tier companies. Recommender Systems make use of many sources of information to provide users with accurate predictions and novel recommendations of items. Here we propose, DeepMF, a novel collaborative fi...
There is a critical issue in explainable recommender systems that compounds the challenges of explainability yet is rarely tackled: the lack of ground-truth explanation texts for training. It is unrealistic to expect every user-item pair in a dataset to have a corresponding target explanation. Hence, we pioneer the first non-supervised explainabili...
Citations
... The framework acquires affiliation administers between areas, eras, and sorts of encounters in view of blog sections. Xue et al. [14] depict an approach for outwardly condensing a milestone by suggesting pictures with various perspectives. It displays a picture's perspective utilizing a 4-D perspective vector, which portrays perspective in level, vertical, scale and introduction angles. ...
The travelogues contain prosperous geo-referenced information such as tours, weather, and expenses etc. Reading travelogues and finding useful trip information is a tedious task for many people. So as to correlate the georeferenced text meaningfully the semantic ranking is taken in account on the user contributing travelogues. In this way there is a requirement for creating programmed travelogue mining methods to pass on valuable data in a travelogue to its peruses in the more successful way. Visualizing a geo-referenced data with location tags and descriptions makes it helpful for readers to understand the fundamental substance of the travelogue. For user expedient, this paper proposes a Travelogue RankTagViz approach that semantically ranks and visualizes the tags of user travelogues. The RankTagViz contains two phases. In the first phase, the user travelogues are ranked based on semantic. In the next phase, the travelogues are visualized based on the tag and images. During semantic ranking, a semantic dimension reduction method is proposed to pre-process and mine useful information from the travelogues. After that a semantic rank mechanism is proposed to rank travelogues based on tags and POI (Point Of Interest). For tag visualization, location based tags and images are extracted from the travelogues and a novel UI is intended to give a superior client encounter, by arranging both the literary and visual data produced by committed voyagers in an alluring way. Test comes about on an arrangement of gathered travelogues show the proposed techniques' capacity to rank and imagine travelogues.
... Content-Based algorithm (Elkahky, 2015;Perkio et al., 2005) and Knowledge-Based algorithm (Kaminskas et al., 2012;Rego et al., 2013) are also applied on the recommendation system. Content-Based recommendation does not need user to evaluate recommendation object (Fan et al., 2012;Tintarev et al., 2011). Instead, it extracts feature of document, analyzes user preference according to user's history and recommend the higher matching document content to the user. ...
... It achieves the state-of-the-art performance in extensive experiments conducted on benchmark datasets for image annotation. Furthermore, we may extend our model by establishing the latent connection between subimages and corresponding tags by means of statistical machine translation [17], [18], [19], or exploring embedding approaches [20], [21], [22], [23], [24] that simultaneously learn the distributed representations for both the image and its tags. ...
Tag-based image retrieval (TBIR) has drawn much attention in recent years due
to the explosive amount of digital images and crowdsourcing tags. However, the
TBIR applications still suffer from the deficient and inaccurate tags provided
by users. Inspired by the subspace clustering methods, we formulate the tag
completion problem in a subspace clustering model which assumes that images are
sampled from subspaces, and complete the tags using the state-of-the-art Low
Rank Representation (LRR) method. And we propose a matrix completion algorithm
to further refine the tags. Our empirical results on multiple benchmark
datasets for image annotation show that the proposed algorithm outperforms
state-of-the-art approaches when handling missing and noisy tags.
... The recommended method in [15] is a graph-based method and Random walk algorithm. The proposed method in [16], first using Trias algorithm [17] discover triadic frequent closed patterns. Then a Hidden Markov Model based on these patterns is made and uses this model in tag recommendation process. ...
With rise of web 2.0, its associated user-centric applications have attracted a lot of users. Folksonomy plays an important role in these systems, which is made of labeling data. Discovery triadic frequent closed patterns is an important tool in knowledge discovery in folksonomy. The huge volume of data and the number of dimensions in these systems, including users, tags and resources are challenging for data mining. In this paper, a method for discovering all triadic frequent closed patterns based on Hidden Markov Model in folksonomy is proposed. By extracting useful data from dataset, the proposed method emprises to build Hidden Markov Model on the two dimensions, then with inference from created hidden model discover triadic frequent closed patterns through applying third dimension on the results. In fact, extracting useful data in the first step and using viterbi based algorithm, for inference, regularly are pruned dataset and are causes for triadic frequent closed patterns to be discovered more quickly. Testing on a real data set taken from 'Del.icio.us' website and comparing the results with the same algorithm in the field of folksonomy called 'Trias' show that the proposed method in terms of the time, can extract all triadic frequent closed patterns more effectively.
In the face of the explosive growth of text information on the Web, people has become increasingly aware of the importance of automatic knowledge mining in use of computers. The knowledge mentioned over this thesis refers to triplets composed of <head_entity, relation, tail_entity>, which convensionally interpret the facts or beliefs in the real world. The research of knowledge mining mainly includes extracting structured knowledge from the free / semi-structured free text, and using the existing structured knowledge to infer undiscovered knowledge. Knowledge extraction and inference methods can help people to automatically extract the core information in massive natural language text and store them in the form of triplets. This makes structured knowledge, whether for information storage and representation, or further enhancing many kinds of applications have broad prospects. For instance, some productized knowledge graph has brought great improvement on user experience and platform performance of many applications on the Internet, such as search engines, QA systems, and even recommender systems.
However, the construction of most large-scale knowledge graphs generally benefits from huge number of users on their platform, in terms of crowdsourcing method to directly obtain knowledge from massive manual editing. The research of automatic knowledge extraction and inference is limited by expensive annotated data, sparse text features, heterogeneous data that is difficult to be consolidated. Recent studies on representation learning theory enlighten us that some characteristics of representation learning may help solve the three problems above. Therefore, this thesis will explore a series of representation learning methods on knowledge extraction and inference which aim at mining new knowledge automatically. The main work and contributions are:
Free text information extraction based on low-rank matrix representation learning. In the aspect of unstructured / semi-structured text, we leverage the assumption of distant supervision to carry out entity linking and automatic construction of weakly labeled training samples for relation extraction, and propose a low-rank matrix completion method to extract relations between entities in text. The highlight of this method is to reconstruct large-scale sparse text features and noises caused by distantly supervised annotation via using the transductive learning method of low-rank matrix completion, while the distant supervision paradigm automatically obtains a large number of weakly annotated data. The principle components of the relationships between text entities represented by a low-rank matrix are learnt simutanously with the noise stripping, so that the performance of relation prediction under distant supervision is improved.
Knowledge representation learning and belief inference based on low-dimensional vectors. In structured knowledge bases, low-dimensional vectors are used to represent and deduce the entities and relations in a knowledge base. We propose geometrical and probabilistic models to conduct the task of belief inference on complete and imperfect knowledge bases, respectively. At the same time, the effect of the multi-relational beliefs that is widely existed in knowledge bases, is analysed and further considered to improve the performance via adapting the learning rate. This method compresses the representation of knowledge from matrix to vector level, not only achieves incremental computation on the algorithm, but also gives the belief inference to the mathematical explaination.
Joint low-dimensional representation learning of knowledge bases and free text. The use of entity descriptions or relation mentions as a link to acquire the low-dimensional vector representations of vocabularies and entities, relationships, bridges the gap between the two kinds of heterogeneous data of free text and knowledge base. The series of approaches proposed not only covers the semantics of entities, relationships, and vocabularies, but also enhances the performance of tasks such as belief inference, entity classification, and relation prediction. Moreover, it shines a light on handling large amount of heterogeneous data for many Web applications such as search engines, QA systems etc.
