DOCSLIB.ORG

Multiple Instance Learning Networks for Fine-Grained Sentiment Analysis

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Multiple Instance Learning Networks for Fine-Grained Sentiment Analysis Multiple Instance Learning Networks for Fine-Grained Sentiment Analysis Stefanos Angelidis and Mirella Lapata Institute for Language, Cognition and Computation School of Informatics, University of Edinburgh 10 Crichton Street, Edinburgh EH8 9AB [email protected], [email protected] Abstract [Rating: ??] I had a very mixed experience at The Stand. The burger and fries were good. The chocolate shake was We consider the task of fine-grained senti- divine: rich and creamy. The drive-thru was horrible. It ment analysis from the perspective of multi- took us at least 30 minutes to order when there were only ple instance learning (MIL). Our neural model is trained on document sentiment labels, and four cars in front of us. We complained about the wait learns to predict the sentiment of text seg- and got a half–hearted apology. I would go back because ments, i.e. sentences or elementary discourse the food is good, but my only hesitation is the wait. units (EDUs), without segment-level supervi- + The burger and fries were good sion. We introduce an attention-based polar- ity scoring method for identifying positive and + The chocolate shake was divine negative text snippets and a new dataset which + I would go back because the food is good we call SPOT (as shorthand for Segment-level – The drive-thru was horrible Summary POlariTy annotations) for evaluating MIL- – It took us at least 30 minutes to order style sentiment models like ours. Experimen- tal results demonstrate superior performance Figure 1: An EDU-based summary of a 2-out-of-5 against multiple baselines, whereas a judge- star review with positive and negative snippets. ment elicitation study shows that EDU-level opinion extraction produces more informative summaries than sentence-based alternatives. Socher et al., 2013) annotated with sentiment la- bels and used to predict sentiment in unseen texts. Coarse-grained document-level annotations are rel- 1 Introduction atively easy to obtain due to the widespread use Sentiment analysis has become a fundamental area of opinion grading interfaces (e.g., star ratings ac- of research in Natural Language Processing thanks companying reviews). In contrast, the acquisition to the proliferation of user-generated content in the of sentence- or phrase-level sentiment labels re- form of online reviews, blogs, internet forums, and mains a laborious and expensive endeavor despite social media. A plethora of methods have been pro- its relevance to various opinion mining applica- posed in the literature that attempt to distill senti- tions, e.g., detecting or summarizing consumer opin- ment information from text, allowing users and ser- ions in online product reviews. The usefulness of vice providers to make opinion-driven decisions. finer-grained sentiment analysis is illustrated in the The success of neural networks in a variety of ap- example of Figure 1, where snippets of opposing po- plications (Bahdanau et al., 2015; Le and Mikolov, larities are extracted from a 2-star restaurant review. 2014; Socher et al., 2013) and the availability of Although, as a whole, the review conveys negative large amounts of labeled data have led to an in- sentiment, aspects of the reviewer’s experience were creased focus on sentiment classification. Super- clearly positive. This goes largely unnoticed when vised models are typically trained on documents focusing solely on the review’s overall rating. (Johnson and Zhang, 2015a; Johnson and Zhang, In this work, we consider the problem of segment- 2015b; Tang et al., 2015; Yang et al., 2016), sen- level sentiment analysis from the perspective of tences (Kim, 2014), or phrases (Socher et al., 2011; Multiple Instance Learning (MIL; Keeler, 1991). 17 Transactions of the Association for Computational Linguistics, vol. 6, pp. 17–31, 2018. Action Editor: Ani Nenkova. Submission batch: 7/17; Revision batch: 11/2017; Published 1/2018. c 2018 Association for Computational Linguistics. Distributed under a CC-BY 4.0 license. Instead of learning from individually labeled seg- larity of a text can be computed (e,g., by aggregating ments, our model only requires document-level su- the sentiment scores of constituent words). More re- pervision and learns to introspectively judge the sen- cently, Taboada et al. (2011) introduced SO-CAL, timent of constituent segments. Beyond showing a state-of-the-art method that combines a rich senti- how to utilize document collections of rated reviews ment lexicon with carefully defined rules over syn- to train fine-grained sentiment predictors, we also tax trees to predict sentence sentiment. investigate the granularity of the extracted segments. Supervised learning techniques have subse- Previous research (Tang et al., 2015; Yang et al., quently dominated the literature (Pang et al., 2002; 2016; Cheng and Lapata, 2016; Nallapati et al., Pang and Lee, 2005; Qu et al., 2010; Xia and 2017) has predominantly viewed documents as se- Zong, 2010; Wang and Manning, 2012; Le and quences of sentences. Inspired by recent work in Mikolov, 2014) thanks to user-generated sentiment summarization (Li et al., 2016) and sentiment clas- labels or large-scale crowd-sourcing efforts (Socher sification (Bhatia et al., 2015), we also represent et al., 2013). Neural network models in particular documents via Rhetorical Structure Theory’s (Mann have achieved state-of-the-art performance on vari- and Thompson, 1988) Elementary Discourse Units ous sentiment classification tasks due to their abil- (EDUs). Although definitions for EDUs vary in the ity to alleviate feature engineering. Kim (2014) literature, we follow standard practice and take the introduced a very successful CNN architecture for elementary units of discourse to be clauses (Carlson sentence-level classification, whereas other work et al., 2003). We employ a state-of-the-art discourse (Socher et al., 2011; Socher et al., 2013) uses recur- parser (Feng and Hirst, 2012) to identify them. sive neural networks to learn sentiment for segments Our contributions in this work are three-fold: of varying granularity (i.e., words, phrases, and sen- a novel multiple instance learning neural model tences). We describe Kim’s (2014) approach in more which utilizes document-level sentiment supervision detail as it is also used as part of our model. to judge the polarity of its constituent segments; the Let xi denote a k-dimensional word embedding creation of SPOT, a publicly available dataset which of the i-th word in text segment s of length n. The contains Segment-level POlariTy annotations (for segment’s input representation is the concatenation sentences and EDUs) and can be used for the eval- of word embeddings x1,..., xn, resulting in word uation of MIL-style models like ours; and the em- matrix X. Let Xi:i+j refer to the concatenation pirical finding (through automatic and human-based of embeddings xi,..., xi+j. A convolution filter evaluation) that neural multiple instance learning is W Rlk, applied to a window of l words, produces ∈ superior to more conventional neural architectures a new feature c = ReLU(W X + b), where i ◦ i:i+l and other baselines on detecting segment sentiment ReLU is the Rectified Linear Unit non-linearity, ‘ ’ 1 ◦ and extracting informative opinions in reviews. denotes the entrywise product followed by a sum over all elements and b R is a bias term. Ap- ∈ 2 Background plying the same filter to every possible window of Our work lies at the intersection of multiple research word vectors in the segment, produces a feature map c = [c1, c2, . , cn l+1]. Multiple feature maps areas, including sentiment classification, opinion − mining and multiple instance learning. We review for varied window sizes are applied, resulting in a related work in these areas below. fixed-size segment representation v via max-over- time pooling. We will refer to the application of con- Sentiment Classification Sentiment classification volution to an input word matrix X, as CNN(X).A is one of the most popular tasks in sentiment anal- final sentiment prediction is produced using a soft- ysis. Early work focused on unsupervised meth- max classifier and the model is trained via back- ods and the creation of sentiment lexicons (Turney, propagation using sentence-level sentiment labels. 2002; Hu and Liu, 2004; Wiebe et al., 2005; Bac- The availability of large-scale datasets (Diao et cianella et al., 2010) based on which the overall po- al., 2014; Tang et al., 2015) has also led to the de- 1Our code and SPOT dataset are publicly available at: velopment of document-level sentiment classifiers https://github.com/stangelid/milnet-sent which exploit hierarchical neural representations. 18 These are obtained by first building representations Opinion Mining A standard setting for opinion of sentences and aggregating those into a document mining and summarization (Lerman et al., 2009; feature vector (Tang et al., 2015). Yang et al. (2016) Carenini et al., 2006; Ganesan et al., 2010; Di Fab- further acknowledge that words and sentences are brizio et al., 2014; Gerani et al., 2014) assumes a set deferentially important in different contexts. They of documents that contain opinions about some en- present a model which learns to attend (Bahdanau et tity of interest (e.g., camera). The goal of the system al., 2015) to individual text parts when constructing is to generate a summary that is representative of the document representations. We describe such an ar- average opinion and speaks to its important aspects chitecture in more detail as we use it as a point of (e.g., picture quality, battery life, value). Output comparison with our own model. summaries can be extractive (Lerman et al., 2009) Given document d comprising segments or abstractive (Gerani et al., 2014; Di Fabbrizio et (s1, . , sm), a Hierarchical Network with at- al., 2014) and the underlying systems exhibit vary- tention (henceforth HIERNET; based on Yang ing degrees of linguistic sophistication from identi- et al., 2016) produces segment representations fying aspects (Lerman et al., 2009) to using RST- (v1,..., vm) which are subsequently fed into a style discourse analysis, and manually defined tem- bidirectional GRU module (Bahdanau et al., 2015), plates (Gerani et al., 2014; Di Fabbrizio et al., 2014).
Load more

Recommended publications
  • Incorporating Automated Feature Engineering Routines Into Automated Machine Learning Pipelines by Wesley Runnels
    Incorporating Automated Feature Engineering Routines into Automated Machine Learning Pipelines by Wesley Runnels Submitted to the Department of Electrical Engineering and Computer Science in partial fulfillment of the requirements for the degree of Master of Engineering in Electrical Engineering and Computer Science at the MASSACHUSETTS INSTITUTE OF TECHNOLOGY May 2020 c Massachusetts Institute of Technology 2020. All rights reserved. Author . Wesley Runnels Department of Electrical Engineering and Computer Science May 12th, 2020 Certified by . Tim Kraska Department of Electrical Engineering and Computer Science Thesis Supervisor Accepted by . Katrina LaCurts Chair, Master of Engineering Thesis Committee 2 Incorporating Automated Feature Engineering Routines into Automated Machine Learning Pipelines by Wesley Runnels Submitted to the Department of Electrical Engineering and Computer Science on May 12, 2020, in partial fulfillment of the requirements for the degree of Master of Engineering in Electrical Engineering and Computer Science Abstract Automating the construction of consistently high-performing machine learning pipelines has remained difficult for researchers, especially given the domain knowl- edge and expertise often necessary for achieving optimal performance on a given dataset. In particular, the task of feature engineering, a key step in achieving high performance for machine learning tasks, is still mostly performed manually by ex- perienced data scientists. In this thesis, building upon the results of prior work in this domain, we present a tool, rl feature eng, which automatically generates promising features for an arbitrary dataset. In particular, this tool is specifically adapted to the requirements of augmenting a more general auto-ML framework. We discuss the performance of this tool in a series of experiments highlighting the various options available for use, and finally discuss its performance when used in conjunction with Alpine Meadow, a general auto-ML package.
    [Show full text]
  • Learning from Few Subjects with Large Amounts of Voice Monitoring Data
    Learning from Few Subjects with Large Amounts of Voice Monitoring Data Jose Javier Gonzalez Ortiz John V. Guttag Robert E. Hillman Daryush D. Mehta Jarrad H. Van Stan Marzyeh Ghassemi Challenges of Many Medical Time Series • Few subjects and large amounts of data → Overfitting to subjects • No obvious mapping from signal to features → Feature engineering is labor intensive • Subject-level labels → In many cases, no good way of getting sample specific annotations 1 Jose Javier Gonzalez Ortiz Challenges of Many Medical Time Series • Few subjects and large amounts of data → Overfitting to subjects Unsupervised feature • No obvious mapping from signal to features extraction → Feature engineering is labor intensive • Subject-level labels Multiple → In many cases, no good way of getting sample Instance specific annotations Learning 2 Jose Javier Gonzalez Ortiz Learning Features • Segment signal into windows • Compute time-frequency representation • Unsupervised feature extraction Conv + BatchNorm + ReLU 128 x 64 Pooling Dense 128 x 64 Upsampling Sigmoid 3 Jose Javier Gonzalez Ortiz Classification Using Multiple Instance Learning • Logistic regression on learned features with subject labels RawRaw LogisticLogistic WaveformWaveform SpectrogramSpectrogram EncoderEncoder RegressionRegression PredictionPrediction •• %% Positive Positive Ours PerPer Window Window PerPer Subject Subject • Aggregate prediction using % positive windows per subject 4 Jose Javier Gonzalez Ortiz Application: Voice Monitoring Data • Voice disorders affect 7% of the US population • Data collected through neck placed accelerometer 1 week = ~4 billion samples ~100 5 Jose Javier Gonzalez Ortiz Results Previous work relied on expert designed features[1] AUC Accuracy Comparable performance Train 0.70 ± 0.05 0.71 ± 0.04 Expert LR without Test 0.68 ± 0.05 0.69 ± 0.04 task-specific feature Train 0.73 ± 0.06 0.72 ± 0.04 engineering! Ours Test 0.69 ± 0.07 0.70 ± 0.05 [1] Marzyeh Ghassemi et al.
    [Show full text]
  • Expert Feature-Engineering Vs. Deep Neural Networks: Which Is Better for Sensor-Free Affect Detection?
    Expert Feature-Engineering vs. Deep Neural Networks: Which is Better for Sensor-Free Affect Detection? Yang Jiang1, Nigel Bosch2, Ryan S. Baker3, Luc Paquette2, Jaclyn Ocumpaugh3, Juliana Ma. Alexandra L. Andres3, Allison L. Moore4, Gautam Biswas4 1 Teachers College, Columbia University, New York, NY, United States [email protected] 2 University of Illinois at Urbana-Champaign, Champaign, IL, United States {pnb, lpaq}@illinois.edu 3 University of Pennsylvania, Philadelphia, PA, United States {rybaker, ojaclyn}@upenn.edu [email protected] 4 Vanderbilt University, Nashville, TN, United States {allison.l.moore, gautam.biswas}@vanderbilt.edu Abstract. The past few years have seen a surge of interest in deep neural net- works. The wide application of deep learning in other domains such as image classification has driven considerable recent interest and efforts in applying these methods in educational domains. However, there is still limited research compar- ing the predictive power of the deep learning approach with the traditional feature engineering approach for common student modeling problems such as sensor- free affect detection. This paper aims to address this gap by presenting a thorough comparison of several deep neural network approaches with a traditional feature engineering approach in the context of affect and behavior modeling. We built detectors of student affective states and behaviors as middle school students learned science in an open-ended learning environment called Betty’s Brain, us- ing both approaches. Overall, we observed a tradeoff where the feature engineer- ing models were better when considering a single optimized threshold (for inter- vention), whereas the deep learning models were better when taking model con- fidence fully into account (for discovery with models analyses).
    [Show full text]
  • Deep Learning Feature Extraction Approach for Hematopoietic Cancer Subtype Classification
    International Journal of Environmental Research and Public Health Article Deep Learning Feature Extraction Approach for Hematopoietic Cancer Subtype Classification Kwang Ho Park 1 , Erdenebileg Batbaatar 1 , Yongjun Piao 2, Nipon Theera-Umpon 3,4,* and Keun Ho Ryu 4,5,6,* 1 Database and Bioinformatics Laboratory, College of Electrical and Computer Engineering, Chungbuk National University, Cheongju 28644, Korea; [email protected] (K.H.P.); [email protected] (E.B.) 2 School of Medicine, Nankai University, Tianjin 300071, China; [email protected] 3 Department of Electrical Engineering, Faculty of Engineering, Chiang Mai University, Chiang Mai 50200, Thailand 4 Biomedical Engineering Institute, Chiang Mai University, Chiang Mai 50200, Thailand 5 Data Science Laboratory, Faculty of Information Technology, Ton Duc Thang University, Ho Chi Minh 700000, Vietnam 6 Department of Computer Science, College of Electrical and Computer Engineering, Chungbuk National University, Cheongju 28644, Korea * Correspondence: [email protected] (N.T.-U.); [email protected] or [email protected] (K.H.R.) Abstract: Hematopoietic cancer is a malignant transformation in immune system cells. Hematopoi- etic cancer is characterized by the cells that are expressed, so it is usually difficult to distinguish its heterogeneities in the hematopoiesis process. Traditional approaches for cancer subtyping use statistical techniques. Furthermore, due to the overfitting problem of small samples, in case of a minor cancer, it does not have enough sample material for building a classification model. Therefore, Citation: Park, K.H.; Batbaatar, E.; we propose not only to build a classification model for five major subtypes using two kinds of losses, Piao, Y.; Theera-Umpon, N.; Ryu, K.H.
    [Show full text]
  • Feature Engineering for Predictive Modeling Using Reinforcement Learning
    The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18) Feature Engineering for Predictive Modeling Using Reinforcement Learning Udayan Khurana Horst Samulowitz Deepak Turaga [email protected] [email protected] [email protected] IBM Research AI IBM Research AI IBM Research AI Abstract Feature engineering is a crucial step in the process of pre- dictive modeling. It involves the transformation of given fea- ture space, typically using mathematical functions, with the objective of reducing the modeling error for a given target. However, there is no well-defined basis for performing effec- tive feature engineering. It involves domain knowledge, in- tuition, and most of all, a lengthy process of trial and error. The human attention involved in overseeing this process sig- nificantly influences the cost of model generation. We present (a) Original data (b) Engineered data. a new framework to automate feature engineering. It is based on performance driven exploration of a transformation graph, Figure 1: Illustration of different representation choices. which systematically and compactly captures the space of given options. A highly efficient exploration strategy is de- rived through reinforcement learning on past examples. rental demand (kaggle.com/c/bike-sharing-demand) in Fig- ure 2(a). Deriving several features (Figure 2(b)) dramatically Introduction reduces the modeling error. For instance, extracting the hour of the day from the given timestamp feature helps to capture Predictive analytics are widely used in support for decision certain trends such as peak versus non-peak demand. Note making across a variety of domains including fraud detec- that certain valuable features are derived through a compo- tion, marketing, drug discovery, advertising, risk manage- sition of multiple simpler functions.
    [Show full text]
  • Explicit Document Modeling Through Weighted Multiple-Instance Learning
    Journal of Artificial Intelligence Research 58 (2017) Submitted 6/16; published 2/17 Explicit Document Modeling through Weighted Multiple-Instance Learning Nikolaos Pappas [email protected] Andrei Popescu-Belis [email protected] Idiap Research Institute, Rue Marconi 19 CH-1920 Martigny, Switzerland Abstract Representing documents is a crucial component in many NLP tasks, for instance pre- dicting aspect ratings in reviews. Previous methods for this task treat documents globally, and do not acknowledge that target categories are often assigned by their authors with gen- erally no indication of the specific sentences that motivate them. To address this issue, we adopt a weakly supervised learning model, which jointly learns to focus on relevant parts of a document according to the context along with a classifier for the target categories. Originated from the weighted multiple-instance regression (MIR) framework, the model learns decomposable document vectors for each individual category and thus overcomes the representational bottleneck in previous methods due to a fixed-length document vec- tor. During prediction, the estimated relevance or saliency weights explicitly capture the contribution of each sentence to the predicted rating, thus offering an explanation of the rating. Our model achieves state-of-the-art performance on multi-aspect sentiment analy- sis, improving over several baselines. Moreover, the predicted saliency weights are close to human estimates obtained by crowdsourcing, and increase the performance of lexical and topical features for review segmentation and summarization. 1. Introduction Many NLP tasks such as document classification, question answering, and summarization heavily rely on how well the contents of the document are represented in a given model.
    [Show full text]
  • Miforests: Multiple-Instance Learning with Randomized Trees⋆
    MIForests: Multiple-Instance Learning with Randomized Trees? Christian Leistner, Amir Saffari, and Horst Bischof Institute for Computer Graphics and Vision, Graz University of Technology, Inffeldgasse 16, 8010 Graz, Austria {leistner,saffari,bischof}@icg.tugraz.at http://www.icg.tugraz.at Abstract. Multiple-instance learning (MIL) allows for training classi- fiers from ambiguously labeled data. In computer vision, this learning paradigm has been recently used in many applications such as object clas- sification, detection and tracking. This paper presents a novel multiple- instance learning algorithm for randomized trees called MIForests. Ran- domized trees are fast, inherently parallel and multi-class and are thus increasingly popular in computer vision. MIForest combine the advan- tages of these classifiers with the flexibility of multiple instance learning. In order to leverage the randomized trees for MIL, we define the hidden class labels inside target bags as random variables. These random vari- ables are optimized by training random forests and using a fast iterative homotopy method for solving the non-convex optimization problem. Ad- ditionally, most previously proposed MIL approaches operate in batch or off-line mode and thus assume access to the entire training set. This limits their applicability in scenarios where the data arrives sequentially and in dynamic environments. We show that MIForests are not limited to off-line problems and present an on-line extension of our approach. In the experiments, we evaluate MIForests on standard visual MIL bench- mark datasets where we achieve state-of-the-art results while being faster than previous approaches and being able to inherently solve multi-class problems.
    [Show full text]
  • A New Modal Autoencoder for Functionally Independent Feature Extraction
    A New Modal Autoencoder for Functionally Independent Feature Extraction Yuzhu Guo Kang Pan Simeng Li Zongchang Han Kexin Wang Department of Automation Sciences and Electrical Engineering Beihang University Beijing, China {yuzhuguo, lkwxl, lsimeng, zchan, wkx933}@buaa.edu.cn Li Li Department of Automation Sciences and Electrical Engineering Beihang University Beijing, China [email protected] Abstract Autoencoders have been widely used for dimensional reduction and feature ex- traction. Various types of autoencoders have been proposed by introducing reg- ularization terms. Most of these regularizations improve representation learning by constraining the weights in the encoder part, which maps input into hidden nodes and affects the generation of features. In this study, we show that a con- straint to the decoder can also significantly improve its performance because the decoder determines how the latent variables contribute to the reconstruction of input. Inspired by the structural modal analysis method in mechanical engineering, a new modal autoencoder (MAE) is proposed by othogonalising the columns of the readout weight matrix. The new regularization helps to disentangle explanatory factors of variation and forces the MAE to extract fundamental modes in data. The learned representations are functionally independent in the reconstruction of input and perform better in consecutive classification tasks. The results were validated on the MNIST variations and USPS classification benchmark suite. Comparative experiments clearly show that the new algorithm has a surprising advantage. The new MAE introduces a very simple training principle for autoencoders and could be promising for the pre-training of deep neural networks. arXiv:2006.14390v1 [cs.LG] 25 Jun 2020 1 Introduction The success of machine learning heavily depends on data representation.
    [Show full text]
  • Feature Engineering for Machine Learning
    Feature Engineering for Machine Learning PRINCIPLES AND TECHNIQUES FOR DATA SCIENTISTS Alice Zheng & Amanda Casari Feature Engineering for Machine Learning Principles and Techniques for Data Scientists Alice Zheng and Amanda Casari Beijing Boston Farnham Sebastopol Tokyo Feature Engineering for Machine Learning by Alice Zheng and Amanda Casari Copyright © 2018 Alice Zheng, Amanda Casari. All rights reserved. Printed in the United States of America. Published by O’Reilly Media, Inc., 1005 Gravenstein Highway North, Sebastopol, CA 95472. O’Reilly books may be purchased for educational, business, or sales promotional use. Online editions are also available for most titles (http://oreilly.com/safari). For more information, contact our corporate/insti‐ tutional sales department: 800-998-9938 or [email protected]. Editors: Rachel Roumeliotis and Jeff Bleiel Indexer: Ellen Troutman Production Editor: Kristen Brown Interior Designer: David Futato Copyeditor: Rachel Head Cover Designer: Karen Montgomery Proofreader: Sonia Saruba Illustrator: Rebecca Demarest April 2018: First Edition Revision History for the First Edition 2018-03-23: First Release See http://oreilly.com/catalog/errata.csp?isbn=9781491953242 for release details. The O’Reilly logo is a registered trademark of O’Reilly Media, Inc. Feature Engineering for Machine Learning, the cover image, and related trade dress are trademarks of O’Reilly Media, Inc. While the publisher and the authors have used good faith efforts to ensure that the information and instructions contained in this work are accurate, the publisher and the authors disclaim all responsibility for errors or omissions, including without limitation responsibility for damages resulting from the use of or reliance on this work. Use of the information and instructions contained in this work is at your own risk.
    [Show full text]
  • A Brief Introduction to Deep Learning
    A Brief Introduction to Deep Learning --Yangyan Li How would you crack it? How to avoid being cracked? Seam Carving! Labradoodle or fried chicken Puppy or bagel Sheepdog or mop Chihuahua or muffin Barn owl or apple Parrot or guacamole Raw chicken or Donald Trump But, we human actually lose! • A demo that shows we, human, lose, on the classification task, we are proud of, we have been trained for millions of years! • If we want to make it hard for bots, it has to be hard for human as well. How would you crack it? We human lose on Go! We (will) lose on many specific tasks! • Speech recognition • Translation • Self-driving • … • BUT, they are not AI yet… • Don’t worry until it dates with your girl/boy friend… Deep learning is so cool for so many problems… A Brief Introduction to Deep Learning • Artificial Neural Network • Back-propagation • Fully Connected Layer • Convolutional Layer • Overfitting Artificial Neural Network 1. Activation function 2. Weights 3. Cost function 4. Learning algorithm Live Demo Neurons are functions Neurons are functions Back-propagation Now, serious stuff, a bit… Fully Connected Layers “When in doubt, use brute force.” --Ken Thompson “If brute force is possible...” --Yangyan Li Convolutional Layers Convolutional Layers Convolution Filters Feature Engineering vs. Learning • Feature engineering is the process of using domain knowledge of the data to create features that make machine learning algorithms work. • “When working on a machine learning problem, feature engineering is manually designing what the input x's should be.” -- Shayne Miel • “Coming up with features is difficult, time- consuming, requires expert knowledge.” --Andrew Ng How to detect it in training process? Dropout Sigmod ReLU Sigmod ReLU Compute, connect, evaluate, correct, train madly… Non-linearity, distributed representation, parallel computation, adaptive, self-organizing… A brief history • McCulloch, Warren S., and Walter Pitts.
    [Show full text]
  • Truncated SVD-Based Feature Engineering for Music Recommendation KKBOX’S Music Recommendation Challenge at ACM WSDM Cup 2018
    Truncated SVD-based Feature Engineering for Music Recommendation KKBOX’s Music Recommendation challenge at ACM WSDM Cup 2018 Nima Shahbazi Mohamed Chahhou Jarek Gryz York University Sidi Mohamed Ben Abdellah York University [email protected] University [email protected] [email protected] ABSTRACT to suggest relevant preference to a specific user. Many web ser- 1 This year’s ACM WSDM Cup asked the ML community to build an vices such as Netflix, Spotify, YouTube, KKBOX use CF to deliver improved music recommendation system using a dataset provided personalized recommendation to their customers. by KKBOX. The task was to predict the chances a user would listen The ACM WSDM Cup challenged the ML community to improve to a song repetitively after the first observable listening event within KKBOX model and build a better music recommendation system 2 a given time frame. We cast this problem as a binary classification using their dataset . KKBOX currently uses CF based algorithm problem and addressed it by using gradient boosted decision trees. with matrix factorization but expects that other techniques could To overcome the cold start problem, which is a notorious problem in lead to better music recommendations. recommender systems, we create truncated SVD-based embedding In this task, we want to predict songs which a user will truly like. features for users, songs and artists. Using the embedding features Intuitively, if a user much enjoys a song, s/he will repetitively listen with four different statistical based features (users, songs, artists to it. We are asked to predict the chances of a user listening to a song and time), our model won the ACM challenge, ranking second.
    [Show full text]
  • A Hybrid Method Based on Extreme Learning Machine and Wavelet Transform Denoising for Stock Prediction
    entropy Article A Hybrid Method Based on Extreme Learning Machine and Wavelet Transform Denoising for Stock Prediction Dingming Wu, Xiaolong Wang * and Shaocong Wu College of Computer Science and Technology, Harbin Institute of Technology, Shenzhen 518055, China; [email protected] (D.W.); [email protected] (S.W.) * Correspondence: [email protected] Abstract: The trend prediction of the stock is a main challenge. Accidental factors often lead to short-term sharp fluctuations in stock markets, deviating from the original normal trend. The short- term fluctuation of stock price has high noise, which is not conducive to the prediction of stock trends. Therefore, we used discrete wavelet transform (DWT)-based denoising to denoise stock data. Denoising the stock data assisted us to eliminate the influences of short-term random events on the continuous trend of the stock. The denoised data showed more stable trend characteris- tics and smoothness. Extreme learning machine (ELM) is one of the effective training algorithms for fully connected single-hidden-layer feedforward neural networks (SLFNs), which possesses the advantages of fast convergence, unique results, and it does not converge to a local minimum. Therefore, this paper proposed a combination of ELM- and DWT-based denoising to predict the trend of stocks. The proposed method was used to predict the trend of 400 stocks in China. The predic- tion results of the proposed method are a good proof of the efficacy of DWT-based denoising for stock trends, and showed an excellent performance compared to 12 machine learning algorithms (e.g., recurrent neural network (RNN) and long short-term memory (LSTM)).
    [Show full text]