5,401 Hits in 5.2 sec

Multi-domain Alias Matching Using Machine Learning

Michael Ashcroft, Fredrik Johansson, Lisa Kaati, Amendra Shrestha
2016 2016 Third European Network Intelligence Conference (ENIC)  
aliases.  ...  While most previous research on social media identity linkage relies on matching usernames, our methodology can also be used for users who actively try to choose dissimilar usernames when creating their  ...  For the alias matching problem, we can distinguish between two main cases: 1) The aliases a 1 and a 2 are used on the same social media site (intra-platform alias matching), or 2) the aliases stem from  ... 
doi:10.1109/enic.2016.019 dblp:conf/enic/AshcroftJKS16 fatcat:irpr3u544revpkqbbk5imbgw5e

Discovery of Drug Brand Names from the Web

Thangaraj M, Sivagaminathan P Ganesan Ganesan
2016 International Journal of Advancements in Technology  
In this paper, drug alias retrieval using regular expression has shown significant improvement in Precision and a fair result for Recall, and F-Score.  ...  Therefore, this method follows semisupervised machine learning technique from known to unknown in extracting brand name aliases online.  ...  The code initially learns through examples as given in the training dataset inducing supervised machine learning.  ... 
doi:10.4172/0976-4860.1000170 fatcat:hm5wvf42afaehm6xbxvsjen6b4

Non-subsampled Contourlet Transform-Based Image Denoising in Ultrasound Images Using Elliptical Directional Windows and Block-Based Noise Estimation [chapter]

J. Jai Jaganath Babu, Gnanou Florence Sudha
2013 Advances in Intelligent Systems and Computing  
Lobiyal Evaluation of English-to-Urdu Machine Translation . . . . . . . . . . . . . 351 Vaishali Gupta, Nisheeth Joshi and Iti Mathur A Novel Edge Detection Technique for Multi-Focus Images Using Image  ...  Appavu alias Balamurugan and S.  ... 
doi:10.1007/978-81-322-1665-0_23 fatcat:ojybq5oj7fahrkpfxlwixukgmu

Disease Named Entity Recognition by Machine Learning Using Semantic Type of Metathesaurus

Huang Zhong, Xiaohua Hu
2013 International Journal of Machine Learning and Computing  
To tackle those problems different approaches have been applied on NER using rule based, dictionary matching based, and machine learning based techniques.  ...  If a word in the sentence is matched with the dictionary entry, it is labeled as a feature for machine learning.  ... 
doi:10.7763/ijmlc.2013.v3.367 fatcat:cmotpq52yzgvxfpocecggea7n4

Investigating the Role of Argumentation in the Rhetorical Analysis of Scientific Publications with Neural Multi-Task Learning Models

Anne Lauscher, Goran Glavaš, Simone Paolo Ponzetto, Kai Eckert
2018 Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing  
By coupling rhetorical classifiers with the extraction of argumentative components in a joint multi-task learning setting, we obtain significant performance gains for different rhetorical analysis tasks  ...  To this end, we (1) augment a corpus of scientific publications annotated with four layers of rhetoric annotations with argumentation annotations and (2) investigate neural multi-task learning architectures  ...  Admittedly, the corpus we used in this work is limited to the domain of computer graphics.  ... 
doi:10.18653/v1/d18-1370 dblp:conf/emnlp/LauscherGP018 fatcat:gyfqgoj5prcynkl7z6yb4oevwy

Record linkage of banks and municipalities through multiple criteria and neural networks

Antonio Maratea, Angelo Ciaramella, Giuseppe Pio Cianci
2020 PeerJ Computer Science  
For validation, seven real datasets have been used (three from banks and four from municipalities), purposely chosen in the same geographical area to increase the probability of matches.  ...  of the available data, first several similarity measures are combined and weighted to build a feature vector, then a Multi-Layer Perceptron (MLP) network is trained and tested to find matching pairs.  ...  Recent literature on Machine Learning applied to RL includes Aiken et al. (2019) , that compares probabilistic, stochastic and machine learning approaches, showing that supervised methods outperform unsupervised  ... 
doi:10.7717/peerj-cs.258 pmid:33816910 pmcid:PMC7924437 fatcat:uhqmfn4zfnck7iuhoxhwmzp2ma

Expense Monitoring System using Data Analytics and Machine Learning Techniques

Navaneethakrishnan P, Sathish G
2020 Zenodo  
Learning algorithm KNN with reasonable accuracy.  ...  of the employee expense through three stage verification process and provides the detailed dynamic analysis of expenses as well as helps employee in knowing the result of his expense priorly through Machine  ...  Machine Learning makes use of data sets and processes these data sets using Machine Learning Algorithms to perform a task without using any explicit instruction.  ... 
doi:10.5281/zenodo.3603611 fatcat:la7k54drnvho7lvjtbcfw5i3ly

Domain-Driven News Representation Using Conditional Attribute-Value Pairs [chapter]

Mihail Minev, Christoph Schommer
2014 Lecture Notes in Computer Science  
In this paper, we present a representation formalism that supports a linguistic composition for machine learning tasks.  ...  Considering announcements in the monetary policy domain, we distinguish between attributes and their values and argue that attributes are to be represented as an aggregated set of economic terms, keeping  ...  With respect to the linguistic annotation, we utilized the OpenNLP machine learning toolkit [22] .  ... 
doi:10.1007/978-3-642-54798-0_11 fatcat:b7sieqkrvvgfnfsqbnue7c4rqu

Recognizing names in biomedical texts using mutual information independence model and SVM plus sigmoid

2006 International Journal of Medical Informatics  
In addition, a support vector machine (SVM) plus sigmoid is proposed to resolve the data sparseness problem in the MIIM.  ...  KEYWORDS Biomedical name recognition; Mutual information independence model; Support vector machine Summary In this paper, we present a biomedical name recognition system, called PowerBioNE.  ...  The main contributions of our work are the novel name alias feature in the biomedical domain, the SVM plus sigmoid approach in the effective resolution of the data sparseness problem in our system and  ... 
doi:10.1016/j.ijmedinf.2005.06.012 pmid:16112894 fatcat:vk2qbpzby5bpjnrdamvp4b3sfq

Knowledge Base Relation Detection via Multi-View Matching [article]

Yang Yu, Kazi Saidul Hasan, Mo Yu, Wei Zhang, Zhiguo Wang
2018 arXiv   pre-print
In this paper, we propose a KB relation detection model via multi-view matching which utilizes more useful information extracted from question and KB.  ...  The matching inside each view is through multiple perspectives to compare two input texts thoroughly. All these components are designed in an end-to-end trainable neural network model.  ...  Due to recent machine learning and especially deep learning advances, many recent proposed RE approaches begin to explore the benefits of deep learning instead of using hand-crafted features.  ... 
arXiv:1803.00612v2 fatcat:cml6vb7ju5euxggx2lfxhxnmgm

Lightly-supervised Representation Learning with Global Interpretability [article]

Marco A. Valenzuela-Escárcega and Ajay Nagesh and Mihai Surdeanu
2018 arXiv   pre-print
Our algorithm iteratively learns custom embeddings for both the multi-word entities to be extracted and the patterns that match them from a few example entities per category.  ...  ., use of limited annotations and interpretability of extraction patterns, with the robust learning approaches proposed in representation learning.  ...  Interpretable models mitigate the technical debt of machine learning (Sculley et al., 2014) . For example, it allows domain experts to make manual, gradual improvements to the models.  ... 
arXiv:1805.11545v1 fatcat:y3codxtxwnfz3epvmawnwg6xfu

InfoXtract: A customizable intermediate level information extraction engine

2006 Natural Language Engineering  
This implies the need for a domain-independent IE system that can easily be customized for specific domains: end users must be given tools to customize the system on their own.  ...  It describes new IE tasks such as entity profiles, and concept-based general events which represent realistic goals in terms of what can be accomplished in the near-term as well as providing useful, actionable  ...  Both supervised machine learning and unsupervised learning are used in InfoXtract.  ... 
doi:10.1017/s1351324906004116 fatcat:f7xidowuffc2bfalikhzdxhw3y

Crossing the Conversational Chasm: A Primer on Natural Language Processing for Multilingual Task-Oriented Dialogue Systems [article]

Evgeniia Razumovskaia, Goran Glavaš, Olga Majewska, Edoardo M. Ponti, Anna Korhonen, Ivan Vulić
2021 arXiv   pre-print
To overcome this limitation, we draw parallels between components of the ToD pipeline and other NLP tasks, which can inspire solutions for learning in low-resource scenarios.  ...  Although this technology represents one of the central objectives of AI and has been the focus of ever more intense research and development efforts, it is currently limited to a few narrow domains (e.g  ...  The best performance in such tasks is obtained with supervised machine learning models.  ... 
arXiv:2104.08570v2 fatcat:bi5xizz4wzct5fpiuk3ikotjta

Multi-step Retriever-Reader Interaction for Scalable Open-domain Question Answering [article]

Rajarshi Das, Shehzaad Dhuliawala, Manzil Zaheer, Andrew McCallum
2019 arXiv   pre-print
Finally, we show that our multi-step-reasoning framework brings consistent improvement when applied to two widely used reader architectures DrQA and BiDAF on various large open-domain datasets --- TriviaQA-unfiltered  ...  The framework is agnostic to the architecture of the machine reading model, only requiring access to the token-level hidden representations of the reader.  ...  The reader is trained using supervised learning (using the correct spans as supervision) and the parameters of the GRU network are trained using reinforcement learning.  ... 
arXiv:1905.05733v1 fatcat:56qbjh5hznhhzchoa4quxvkoke

What's in a Name? Answer Equivalence For Open-Domain Question Answering [article]

Chenglei Si, Chen Zhao, Jordan Boyd-Graber
2021 arXiv   pre-print
This work explores mining alias entities from knowledge bases and using them as additional gold answers (i.e., equivalent answers).  ...  Answer expansion increases the exact match score on all datasets for evaluation, while incorporating it helps model training over real-world datasets.  ...  We use batch size of 16 and learning rate of 3e-5 for training on all datasets. Augmented Evaluation.  ... 
arXiv:2109.05289v1 fatcat:2mgtnovvlvd2hfl7raeiqvxeqq
« Previous Showing results 1 — 15 out of 5,401 results