Filters








144 Hits in 3.0 sec

Efficient topic-based unsupervised name disambiguation

Yang Song, Jian Huang, Isaac G. Councill, Jia Li, C. Lee Giles
2007 Proceedings of the 2007 conference on Digital libraries - JCDL '07  
Our models explicitly introduce a new variable for persons and learn the distribution of topics with regard to persons and words.  ...  Experiments on web data and scientific documents from CiteSeer indicate that our approach consistently outperforms other unsupervised learning methods such as spectral clustering and DBSCAN clustering  ...  Our Contribution The objective of this paper is to propose an approach of name disambiguation that includes the attractive properties of both supervised and unsupervised learning methods while trying to  ... 
doi:10.1145/1255175.1255243 dblp:conf/jcdl/SongHCLG07 fatcat:k26gwgsok5cqnas7uapu2obzhy

Weakly Supervised Domain-Specific Color Naming Based on Attention [article]

Lu Yu, Yongmei Cheng, Joost van de Weijer
2018 arXiv   pre-print
However, in many applications, different sets of color names are used for the accurate description of objects.  ...  Labeling data to learn these domain-specific color names is an expensive and laborious task. Therefore, in this article we aim to learn color names from weakly labeled data.  ...  We abbreviate attention, centric information and alternating learning as AM, C, AL. Accuracy Ours 55.45 Ours+AM 84.09 Ours+AM+C 84.77 Ours+AM+C+AL 86.59 Fig. 7.  ... 
arXiv:1805.04385v1 fatcat:hkjj75vx25gxdefylu5e63kxcm

Recognition of Activities of Daily Living from Topic Model

Isibor Kennedy Ihianle, Usman Naeem, Abdel-Rahman Tawil
2016 Procedia Computer Science  
The data captured from these sensors are required to be classified using various machine learning or knowledge driven techniques to infer and recognise activities.  ...  The process of discovering the activities and activity-object patterns from the sensors tagged to objects as they are used is critical to recognising the activities.  ...  be handled and the number of objects capable of being tagged with them.  ... 
doi:10.1016/j.procs.2016.09.007 fatcat:ducjjyhvbjdm3n6sghgbn6cobu

Geo-located image analysis using latent representations

M. Cristani, A. Perina, U. Castellani, V. Murino
2008 2008 IEEE Conference on Computer Vision and Pattern Recognition  
The proposed strategy permits also to deal with the geo-recognition problem, i.e., to infer the geographical area depicted by images with no available location information.  ...  The method lies in the wide literature on statistical latent representations, in particular, the probabilistic Latent Semantic Analysis (pLSA) paradigm has been extended, introducing a latent aspect which  ...  community for unsupervised topic discovery in a corpus of documents.  ... 
doi:10.1109/cvpr.2008.4587390 dblp:conf/cvpr/CristaniPCM08 fatcat:gbdzir56a5e4horow6zjwuxjs4

A collective topic model for milestone paper discovery

Ziyu Lu, Nikos Mamoulis, David W. Cheung
2014 Proceedings of the 37th international ACM SIGIR conference on Research & development in information retrieval - SIGIR '14  
We propose a collective topic model on three types of objects: papers, authors and published venues. We model any of these objects as bags of citations.  ...  Based on Probabilistic latent semantic analysis (PLSA), authorship, published venues and citation relations are used for quantifying paper importance.  ...  We developed our model based on PLSA [4] .  ... 
doi:10.1145/2600428.2609499 dblp:conf/sigir/LuMC14 fatcat:ag4cco2zbjc5nn2qwefikinz4i

SAR-Based Terrain Classification Using Weakly Supervised Hierarchical Markov Aspect Models

Wen Yang, Dengxin Dai, Bill Triggs, Gui-Song Xia
2012 IEEE Transactions on Image Processing  
Index Terms-Hierarchical Markov aspect model (HMAM), probabilistic latent semantic analysis (PLSA), scene labeling, synthetic aperture radar.  ...  single-scale aspect models with only a modest increase in training and test complexity.  ...  learned for each level.  ... 
doi:10.1109/tip.2012.2199127 pmid:22614643 fatcat:nujxbwqbsfga5edlgctdfyuv6u

AttentionBased Deep Feature Fusion for the Scene Classification of HighResolution Remote Sensing Images

Zhu, Yan, Mo, Liu
2019 Remote Sensing  
The experiments confirm that the proposed method outperforms most competitive scene classification methods with an average overall accuracy of 94% under different training ratios.  ...  Deep learning techniques, especially the convolutional neural network (CNN) have been widely applied to the scene classification of HRRSI due to the advancement of graphic processing units (GPU).  ...  of the ADFF framework compared with the unsupervised feature learning methods.  ... 
doi:10.3390/rs11171996 fatcat:fld3bspflfbmhhhzt2c5xe3htq

Constructing Topic Models of Internet of Things for Information Processing

Jie Xin, Zhiming Cui, Shukui Zhang, Tianxu He, Chunhua Li, Haojing Huang
2014 The Scientific World Journal  
There is abundant digital products data on the IoT, linking with multiple types of objects/entities. Those associated entities carry rich information and usually in the form of query records.  ...  In this paper, we propose a novelrecord entity topic model(RETM) for IoT environment that is associated with a set of entities and records and a Gibbs sampling-based algorithm is proposed to learn the  ...  The concept behind IoT is that a variety of objects around us can interact and work with each other to pursue common goals [1] .  ... 
doi:10.1155/2014/675234 pmid:25110737 pmcid:PMC4119721 fatcat:4hxligtscnhdtjc7uibqyqdqom

A SURVEY ON SENTIMENT ANALYSIS

Madhusudhanan S., Dr. Moorthi M.
2018 Indian Journal of Computer Science and Engineering  
I am very happy with this camera so far.I love how compact it is and the fun colors it comes in."  ...  Types of Machine learning are (1) Supervised Learning and (2) Unsupervised Learning.Summarizes of recent work related to opinion mining are shown in Table. 1. 2.4.1Supervised learning : This requires  ... 
doi:10.21817/indjcse/2018/v9i2/180902030 fatcat:rotwyyktrfephjq2i53q4jspoi

A Discriminative Representation of Convolutional Features for Indoor Scene Recognition

Salman H. Khan, Munawar Hayat, Mohammed Bennamoun, Roberto Togneri, Ferdous A. Sohel
2016 IEEE Transactions on Image Processing  
To this end, we introduce a new large-scale dataset of 1300 object categories which are commonly present in indoor scenes.  ...  Traditionally used convolutional features preserve the global spatial structure, which is a desirable property for general object recognition.  ...  (b) Image of a "Library", a "Museum" and a "Church" (left to right): How can we hope to learn the subtle differences between different scene types? Fig. 1 : 'Where am I located indoors?'  ... 
doi:10.1109/tip.2016.2567076 pmid:28113718 fatcat:u2kt4yc55vb3rapc2eqyvyrssq

Auto-Dialabel: Labeling Dialogue Data with Unsupervised Learning

Chen Shi, Qi Chen, Lei Sha, Sujian Li, Xu Sun, Houfeng Wang, Lintao Zhang
2018 Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing  
Moreover, since the whole process is unsupervised, it can be much faster and more consistent and objective than human labeling, and extended to other domains.  ...  S t is also learned automatically and dynamically.  ... 
doi:10.18653/v1/d18-1072 dblp:conf/emnlp/ShiCSLSWZ18 fatcat:w466ygbbgvg4ba77crwzbt35ky

Opinion Aspects Based on Customer Feelings via Reviews

T Sajana, Hanuman .
2018 International Journal of Engineering & Technology  
More mind-boggling calculations are utilized to address this issue with expansive datasets.  ...  Article acquaints an approach with perceive and condense item perspectives and concentrate sentiments from an immense number of item surveys in an area.  ...  They are: (1) Unsupervised learning technique and (2) Supervised Machine Learning.  ... 
doi:10.14419/ijet.v7i3.12.17873 fatcat:rmkmljcsizczhicg5rdiuqucwi

Multi-task Semi-supervised Semantic Feature Learning for Classification

Changying Du, Fuzhen Zhuang, Qing He, Zhongzhi Shi
2012 2012 IEEE 12th International Conference on Data Mining  
labeled information into traditional unsupervised learning of latent semantics.  ...  Finally extensive experiments on text data along with the comparison with various baselines and three state-of-the-art multi-task learning algorithms demonstrate the effectiveness of our method.  ...  Our objective is to learn these tasks together with the hope that they will help each other in the learning process.  ... 
doi:10.1109/icdm.2012.15 dblp:conf/icdm/DuZHS12 fatcat:vfy4azkvvbcnni3mi43cq56u2e

EXPLORING INFORMATION RETRIEVAL BY LATENT SEMANTIC INDEXING AND LATENT DIRICHLET ALLOCATION TECHNIQUES

Radha Guha
2020 International Research Journal of Computer Science  
This paper explores information retrieval models and experiments Semantic Indexing (LSI) first and then with the more efficient topic modeling algorithm of Latent Dirichlet Allocation (LDA).  ...  LDA is an unsupervised learning model, which generates soft clusters of topics, which overlap as a participating word can belong to more than one topic because of polysemy of English language.  ...  For example in lemmatization "am", "are", and "is" are converted to "be".  ... 
doi:10.26562/irjcs.2020.v0705.001 fatcat:3mmmcy5kuve5hetxfh456bxwoy

Mining multi-faceted overviews of arbitrary topics in a text collection

Xu Ling, Qiaozhu Mei, ChengXiang Zhai, Bruce Schatz
2008 Proceeding of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD 08  
to mine a multi-faceted overview in an unsupervised way.  ...  by supervised methods with training examples.  ...  I am very pleased with fuel economy.  ... 
doi:10.1145/1401890.1401952 dblp:conf/kdd/LingMZS08 fatcat:mz3bnsgkg5buxf5ku5mrltomye
« Previous Showing results 1 — 15 out of 144 results