scholarly journals Analysing Predictive Coding Algorithms for Document Review

2021 ◽  
Vol 9 (11) ◽  
pp. 1679-1681
Author(s):  
Aditi Wikhe
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Text Classification ◽  
Unscented Kalman Filter ◽  
Predictive Coding ◽  
Machine Learning Techniques ◽  
Learning Models ◽  
Legal Domain ◽  
The Right ◽  
Document Review

Abstract: Lawsuits and regulatory investigations in today's legal environment demand corporations to engage in increasingly intense data-focused engagements to find, acquire, and evaluate vast amounts of data. In recent years, technology-assisted review (TAR) has become a more crucial part of the document review process in legal discovery. Attorneys now have been using machine learning techniques like text classification to identify responsive information. In the legal domain, text classification is referred to as predictive coding or technology assisted review (TAR). Predictive coding is used to increase the number of relevant documents identified, while reducing human labelling efforts and manual review of documents. Deep learning models mixed with word embeddings have demonstrated to be more effective in predictive coding in recent years. Deep learning models, on the other hand, have a lot of variables, making it difficult and time-consuming for legal professionals to choose the right settings. In this paper, we will look at a few predictive coding algorithms and discuss which one is the most efficient among them. Keywords: Technology-assisted-review, predictive coding, machine learning, text classification, deep learning, CNN , Unscented Kalman Filter, Logistic Regression, SVM

scholarly journals The Unreasonable Effectiveness of the Baseline: Discussing SVMs in Legal Text Classification

2021 ◽  
Author(s):  
Benjamin Clavié ◽  
Marc Alphonsus
Keyword(s):  
Deep Learning ◽  
Language Processing ◽  
Text Classification ◽  
Traditional Approach ◽  
Error Reduction ◽  
Support Vector ◽  
Learning Models ◽  
Legal Text ◽  
Classification Tasks ◽  
Legal Domain

We aim to highlight an interesting trend to contribute to the ongoing debate around advances within legal Natural Language Processing. Recently, the focus for most legal text classification tasks has shifted towards large pre-trained deep learning models such as BERT. In this paper, we show that a more traditional approach based on Support Vector Machine classifiers reaches competitive performance with deep learning models. We also highlight that error reduction obtained by using specialised BERT-based models over baselines is noticeably smaller in the legal domain when compared to general language tasks. We discuss some hypotheses for these results to support future discussions.

scholarly journals A systematic review of text classification research based on deep learning models in Arabic language

2020 ◽  
Vol 10 (6) ◽  
pp. 6629
Author(s):  
Ahlam Wahdan ◽  
Sendeyah AL Hantoobi ◽  
Said A. Salloum ◽  
Khaled Shaalan
Keyword(s):  
Neural Network ◽  
Systematic Review ◽  
Neural Networks ◽  
Deep Learning ◽  
Text Classification ◽  
Arabic Language ◽  
Machine Learning Techniques ◽  
Learning Models ◽  
Learning Techniques

Classifying or categorizing texts is the process by which documents are classified into groups by subject, title, author, etc. This paper undertakes a systematic review of the latest research in the field of the classification of Arabic texts. Several machine learning techniques can be used for text classification, but we have focused only on the recent trend of neural network algorithms. In this paper, the concept of classifying texts and classification processes are reviewed. Deep learning techniques in classification and its type are discussed in this paper as well. Neural networks of various types, namely, RNN, CNN, FFNN, and LSTM, are identified as the subject of study. Through systematic study, 12 research papers related to the field of the classification of Arabic texts using neural networks are obtained: for each paper the methodology for each type of neural network and the accuracy ration for each type is determined. The evaluation criteria used in the algorithms of different neural network types and how they play a large role in the highly accurate classification of Arabic texts are discussed. Our results provide some findings regarding how deep learning models can be used to improve text classification research in Arabic language.

scholarly journals Comparison of Deep Transfer Learning Techniques in Human Skin Burns Discrimination

2020 ◽  
Vol 3 (2) ◽  
pp. 20 ◽  
Author(s):  
Aliyu Abubakar ◽  
Mohammed Ajuji ◽  
Ibrahim Usman Yahya
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Transfer Learning ◽  
Fine Tuning ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Learning Models ◽  
Skin Injuries ◽  
Learning Techniques ◽  
Injured Skin

While visual assessment is the standard technique for burn evaluation, computer-aided diagnosis is increasingly sought due to high number of incidences globally. Patients are increasingly facing challenges which are not limited to shortage of experienced clinicians, lack of accessibility to healthcare facilities and high diagnostic cost. Certain number of studies were proposed in discriminating burn and healthy skin using machine learning leaving a huge and important gap unaddressed; whether burns and related skin injuries can be effectively discriminated using machine learning techniques. Therefore, we specifically use transfer learning by leveraging pre-trained deep learning models due to deficient dataset in this paper, to discriminate two classes of skin injuries—burnt skin and injured skin. Experiments were extensively conducted using three state-of-the-art pre-trained deep learning models that includes ResNet50, ResNet101 and ResNet152 for image patterns extraction via two transfer learning strategies—fine-tuning approach where dense and classification layers were modified and trained with features extracted by base layers and in the second approach support vector machine (SVM) was used to replace top-layers of the pre-trained models, trained using off-the-shelf features from the base layers. Our proposed approach records near perfect classification accuracy in categorizing burnt skin ad injured skin of approximately 99.9%.

scholarly journals Comparison of Deep Transfer Learning Techniques in Human Skin Burns Discrimination

2020 ◽  
Author(s):  
Aliyu Abubakar ◽  
Mohammed Ajuji ◽  
Ibrahim Usman Yahya
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Learning Strategies ◽  
Transfer Learning ◽  
Standard Technique ◽  
Fine Tuning ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Learning Models ◽  
Learning Techniques

While visual assessment is the standard technique for burn evaluation, computer-aided diagnosis is increasingly sought due to high number of incidences globally. Patients are increasingly facing challenges which are not limited to shortage of experienced clinicians, lack of accessibility to healthcare facilities, and high diagnostic cost. Certain number of studies were proposed in discriminating burn and healthy skin using machine learning leaving a huge and important gap unaddressed; whether burns and related skin injuries can be effectively discriminated using machine learning techniques. Therefore, we specifically use pre-trained deep learning models due to deficient dataset to train a new model from scratch. Experiments were extensively conducted using three state-of-the-art pre-trained deep learning models that includes ResNet50, ResNet101 and ResNet152 for image patterns extraction via two transfer learning strategies: fine-tuning approach where dense and classification layers were modified and trained with features extracted by base layers, and in the second approach support vector machine (SVM) was used to replace top-layers of the pre-trained models, trained using off-the-shelf features from the base layers. Our proposed approach records near perfect classification accuracy of approximately 99.9%.

scholarly journals Deep Learning-Based Approaches for Decoding Motor Intent From Peripheral Nerve Signals

2021 ◽  
Vol 15 ◽  
Author(s):  
Diu K. Luu ◽  
Anh T. Nguyen ◽  
Ming Jiang ◽  
Jian Xu ◽  
Markus W. Drealan ◽  
...  
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Real Time ◽  
Input Data ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Learning Models ◽  
Data Set ◽  
Advantages And Disadvantages ◽  
Trade Offs

Previous literature shows that deep learning is an effective tool to decode the motor intent from neural signals obtained from different parts of the nervous system. However, deep neural networks are often computationally complex and not feasible to work in real-time. Here we investigate different approaches' advantages and disadvantages to enhance the deep learning-based motor decoding paradigm's efficiency and inform its future implementation in real-time. Our data are recorded from the amputee's residual peripheral nerves. While the primary analysis is offline, the nerve data is cut using a sliding window to create a “pseudo-online” dataset that resembles the conditions in a real-time paradigm. First, a comprehensive collection of feature extraction techniques is applied to reduce the input data dimensionality, which later helps substantially lower the motor decoder's complexity, making it feasible for translation to a real-time paradigm. Next, we investigate two different strategies for deploying deep learning models: a one-step (1S) approach when big input data are available and a two-step (2S) when input data are limited. This research predicts five individual finger movements and four combinations of the fingers. The 1S approach using a recurrent neural network (RNN) to concurrently predict all fingers' trajectories generally gives better prediction results than all the machine learning algorithms that do the same task. This result reaffirms that deep learning is more advantageous than classic machine learning methods for handling a large dataset. However, when training on a smaller input data set in the 2S approach, which includes a classification stage to identify active fingers before predicting their trajectories, machine learning techniques offer a simpler implementation while ensuring comparably good decoding outcomes to the deep learning ones. In the classification step, either machine learning or deep learning models achieve the accuracy and F1 score of 0.99. Thanks to the classification step, in the regression step, both types of models result in a comparable mean squared error (MSE) and variance accounted for (VAF) scores as those of the 1S approach. Our study outlines the trade-offs to inform the future implementation of real-time, low-latency, and high accuracy deep learning-based motor decoder for clinical applications.

scholarly journals Music Genre Classification Using Deep Learning with KNN

2021 ◽  
pp. 224-230
Author(s):  
Dr. S. Ponlatha ◽  
Mathisalini B ◽  
Deepthisri K. A ◽  
Kalaiyarasi. M ◽  
Kowshika. V
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Machine Learning Techniques ◽  
Music Score ◽  
Genre Classification ◽  
Learning Techniques ◽  
Music Genre ◽  
The Right ◽  
Music Streaming ◽  
Music Genre Classification

Music genre is a conventional category that predicts the genre of music belonging to tradition or set of conventions. A music platform, with total assets of $26 billion, is ruling the music streaming stage today. At present, it has a huge number of tunes and it is information base and claims to have the right music score for everybody. Like, Spotify, Amazon music, Wynk has put a great deal in examination to further develop the manner in which clients find and pay attention to music. AI is at the centre of their examination. From NLP to Collaborative sifting to Deep Learning, All music platforms utilizes them all. Tunes are examined dependent on their advanced marks for certain elements, including rhythm, acoustics, energy, danceability, and so forth, to answer that incomprehensible old first-date inquiry. Organizations these days use music arrangement, either to have the option to put suggestions to their clients (like Spotify, Soundcloud) or just as an item (for instance, Shazam). Deciding music sorts is the initial phase toward that path. AI procedures have ended up being very fruitful in removing patterns and examples from a huge information pool. Similar standards are applied in Music Analysis moreover. Machine learning techniques are achieved in some recent years and rarely in deep learning. Most of the current music genre classification uses Machine learning techniques. In this, we present a music dataset which includes many genres like Rock, Pop, folk, Classical and many genres. A Deep learning approach is used in order to train and classify the system using KNN.

scholarly journals Machine Learning Techniques for Diagnosis of Lower Gastrointestinal Cancer: A Systematic Review

2021 ◽  
Keyword(s):  
Machine Learning ◽  
Systematic Review ◽  
Literature Review ◽  
Systematic Literature Review ◽  
Gastrointestinal Cancer ◽  
Medical Science ◽  
Learning Model ◽  
Machine Learning Techniques ◽  
Learning Models ◽  
The Right

Background: Nowadays, it can be seen that changes have taken place in the process of diseases and their clinical parameters. Accordingly, in some cases, general medical science and the use of clinical statistics based on the experiences of the physicians are not enough for the provision of sufficient tools for an early and accurate diagnosis. Therefore, medical science increasingly seeks to use unconventional methods and machine learning techniques. The issue of diagnosis in the medical world and the error rate of physicians in this regard are among the main challenges of the condition of patients and diseases. For this reason, in recent years, artificial intelligence tools have been used to help physicians. However, one of the main problems is that the effectiveness of machine learning tools is not studied much. Due to the sensitivity and high prevalence of diseases, especially gastrointestinal cancer, there is a need for a systematic review to identify methods of machine learning and artificial intelligence and compare their impact on the diagnosis of lower gastrointestinal cancers. Objectives: This systematic review aimed to identify the machine learning methods used for the diagnosis of lower gastrointestinal cancers. Moreover, it aimed to classify the presented methods and compare their effectiveness and evaluation indicators. Methods: This systematic review was conducted using six databases. The systematic literature review follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement for systematic reviews. The search strategy consisted of four expressions, namely “machine learning algorithm”, “lower gastrointestinal”, “cancer”, and “diagnosis and screening”, in that order. It should be mentioned that studies based on treatment were excluded from this review. Similarly, studies that presented guidelines, protocols, and instructions were excluded since they only require the focus of clinicians and do not provide progression along an active chain of reasoning. Finally, studies were excluded if they had not undergone a peer-review process. The following aspects were extracted from each article: authors, year, country, machine learning model and algorithm, sample size, the type of data, and the results of the model. The selected studies were classified based on three criteria: 1) machine learning model, 2) cancer type, and 3) effect of machine learning on cancer diagnosis. Results: In total, 44 studies were included in this systematic literature review. The earliest article was published in 2010, and the most recent was from 2019. Among the studies reviewed in this systematic review, one study was performed on the rectum (rectal cancer), one was about the small bowel (small bowel cancer), and 42 studies were on the colon (colon cancer, colorectal cancer, and colonic polyps). In total, 19 out of the 44 (43%) articles from the systematic literature review presented a deep learning model, and 25 (57%) articles used classic machine learning. The models worked mostly on image and all of them were supervised learning models. All studies with deep learning models used Convolutional Neural Network and were published between 2016 and 2019. The studies with classic machine learning models used diverse methods, mostly Support Vector Machine, K-Nearest Neighbors, and Artificial Neural Network. Conclusion: Machine learning methods are suitable tools in the field of cancer diagnosis, especially in cases related to the lower gastrointestinal tract. These methods can not only increase the accuracy of diagnosis and help the doctor to make the right decision, but also help in the early diagnosis of cancer and reduce treatment costs. The methods presented so far have focused more on image data and more than anything else have helped to increase the accuracy of physicians in making the correct diagnosis. Achievement of the right method for early diagnosis requires more accurate data sets and analyses.

Sign in / Sign up

Export Citation Format

Share Document