Self-Refine Learning For Data-Centric Text Classification

<div> <div> <p>In industry NLP application, our manually labeled data has a certain number of noisy data. We present a simple method to find the noisy data and re-label their labels to the result of model prediction. We select the noisy data whose human label is not contained in the top-K model’s predictions. The model is trained on the origin dataset. The experiment result shows that our method works. For industry deep learning application, our method improve the text classification accuracy from 80.5% to 90.6% in dev dataset, and improve the human-evaluation accuracy from 83.2% to 90.5%.<br></p> </div> </div>

Self-Refine Learning For Data-Centric Text Classification

<div> <div> <p>In industry NLP application, our manually labeled data has a certain number of noisy data. We present a simple method to find the noisy data and re-label their labels to the result of model prediction. We select the noisy data whose human label is not contained in the top-K model’s predictions. The model is trained on the origin dataset. The experiment result shows that our method works. For industry deep learning application, our method improve the text classification accuracy from 80.5% to 90.6% in dev dataset, and improve the human-evaluation accuracy from 83.2% to 90.5%.<br></p> </div> </div>

Automatic Rain Detection System Based on Digital Images of CCTV Cameras Using Convolutional Neural Network Method

The Meteorology, Climatology and Geophysics Agency (BMKG) has a duty to provide weather information including rainfall. BMKG has several types of rainfall gauges, but these are not evenly distributed across regions. The solution to increase the density of rainfall observations is to use existing sources to obtain weather information. This research uses Closed Circuit Television (CCTV) that is spread across the Jakarta area to produce information on rainy conditions. The method used is the Convolutional Neural Network (CNN). The image from CCTV will be used for the training and testing process, so as to get the best accuracy model. The results of this model will be used for rain detection on CCTV digital images. The rain detection process is carried out automatically and in real time. The results of the rain detection process will be displayed on the map according to the location where the CCTV was installed. This research has succeeded in making a CNN model for rain detection with a training accuracy of 98.8% and a testing accuracy of 96.4%, as well as evaluating the BMKG observation data, so it has an evaluation accuracy of 96.7%.

Self-Refine Learning For Data-Centric Text Classification

<div> <div> <p>In industry NLP application, our manually labeled data has a certain number of noisy data. We present a simple method to find the noisy data and re-label their labels to the result of model prediction. We select the noisy data whose human label is not contained in the top-K model’s predictions. The model is trained on the origin dataset. The experiment result shows that our method works. For industry deep learning application, our method improve the text classification accuracy from 80.5% to 90.6% in dev dataset, and improve the human-evaluation accuracy from 83.2% to 90.5%.<br></p> </div> </div>

Self-Refine Learning For Data-Centric Text Classification

<div> <div> <p>In industry NLP application, our manually labeled data has a certain number of noisy data. We present a simple method to find the noisy data and re-label their labels to the result of model prediction. We select the noisy data whose human label is not contained in the top-K model’s predictions. The model is trained on the origin dataset. The experiment result shows that our method works. For industry deep learning application, our method improve the text classification accuracy from 80.5% to 90.6% in dev dataset, and improve the human-evaluation accuracy from 83.2% to 90.5%.<br></p> </div> </div>

Evaluation Method of Resource Development Effect of Cultural Industry Based on Wireless Sensor Technology

To overcome the problems of low evaluation accuracy and efficiency existing in traditional resource development efficiency evaluation methods, this paper proposes a new evaluation method of cultural industry tourism resource development effect based on wireless sensor technology. The method takes into account the data of cultural industry resource development collected by the wireless sensors and transmits it to the computer system by wireless transmission technology. Based on the collected development data, the evaluation index system is constructed and the index weight is calculated. Furthermore, the multiobjective weighted function is employed to calculate the development effect evaluation score and complete the evaluation of cultural industry resources development effect. Experimental results show that the proposed method can successfully improve the evaluation accuracy and efficiency, and the highest evaluation accuracy is stable at 98%. Therefore, this method has high reliability.

Enhancing Cyberweapon Effectiveness Methodology With SE Modeling Techniques

A recent cyberweapons effectiveness methodology clearly provides a parallel but distinct process from that of kinetic weapons – both for defense and offense purposes. This methodology promotes consistency and improves cyberweapon system evaluation accuracy – for both offensive and defensive postures. However, integrating this cyberweapons effectiveness methodology into the design phase and operations phase of weapons systems development is still a challenge. The paper explores several systems engineering modeling techniques (e.g., SysML) and how they can be leveraged towards an enhanced effectiveness methodology. It highlights how failure mode analyses (e.g., FMEA) can facilitate cyber damage determination and target assessment, how block and parametric diagraming techniques can facilitate characterizing cyberweapons and eventually assess the effectiveness of such weapons and conversely assess vulnerabilities of systems to certain types of cyberweapons.

ADHD Prediction via Time Series Ensemble fed Driving Simulator Data

In this paper, we identify the on-road scenarios within a simulated driving environment where a group of clinical trial participants (n= 30) with and without Attention Deficit Hyper-activity Disorder (ADHD) drive perceivably different fromone another. We partition the simulated routes into smaller non-overlapping sections in order to determine which sections elicit behaviors that are predictive of ADHD. Then, we develop section-specific classifiers, which are used as voters in bagging ensemble classifiers. Our results show gains in classifying ADHD (increase in 5-fold average evaluation accuracy) over our previous efforts, as well as providing explainable evidence that driving behaviors indicative of ADHD tend to be exhibited in turns and curves.

Quantifying and Evaluating the Cultivated Areas Suitable for Fallow in Chongqing of China Using Multisource Data

The quantitative evaluation of the suitability of land fallow is of great significance to the effective implementation of fallow system in rural China. The purpose of this study is to systematically evaluate the cultivated areas suitable for fallow in Chongqing, China. The results show that: (1) a comprehensive index of cultivated land fallow (ILF) was developed by employing a series of multi—source data, and the ILF has been proven as an effective proxy to identify the cultivated areas suitable for fallow; (2) cultivated land with ILF values above the average value accounts for 34.38% (9902 km2) of the total cultivated land; (3) the ILF is negatively correlated with the population density, transportation proximity, and proportion of inclined area. This study argued that the ILF can reflect the cultivated areas suitable for fallow in Chongqing and can provide guidance for the spatial distribution of cultivated land fallow. The findings indicated that the differences in geographical elements between karst and non—karst areas must be further investigated, and the evaluation accuracy of the cultivated areas suitable for fallow must be improved.