12fab62417720fa68847154 20210805050600206 beie:beie director@blueeyesintelligence.org WEB-FORM International Journal of Recent Technology and Engineering IJRTE 22773878 10.35940/ijrte.2277-3878 https://www.ijrte.org/ 07 30 2021 10 2 10.35940/ijrte.10.2 https://www.ijrte.org/download/volume-10-issue-2/ Professor, Department of School of Computer Science and Engineering, Vellore Institute of Technology, Vellore (Tamil Nadu), India. Aju D Dibyajyoti Roy Department of School of Electronics and Engineering, Vellore Institute of Technology, Vellore (Tamil Nadu), India. Raghav Agarwal Department of School of Computer Science and Engineering, Vellore Institute of Technology, Vellore (Tamil Nadu), India. Tanishq Nagpal Department of School of Computer Science and Engineering, Vellore Institute of Technology, Vellore (Tamil Nadu), India. The high-volume and velocity data stream generated from devices and applications from different domains grows steadily and is valuable for big data research. One of the most important topics is anomaly detection for streaming data, which has attracted attention and investigation in plenty of areas, e.g., the sensor data anomaly detection, predictive maintenance, event detection. Those efforts could potentially avoid large amount of financial costs in the manufacture. However, different from traditional anomaly detection tasks, anomaly detection in streaming data is especially difficult due to that data arrives along with the time with latent distribution changes, so that a single stationary model doesn’t fit streaming data all the time. An anomaly could become normal during the data evolution, therefore it is necessary to maintain a dynamic system to adapt the changes. In this work, we propose a LSTMs-Autoencoder anomaly detection model for streaming data. This is a mini-batch based streaming processing approach. We experimented with streaming data that containing different kinds of anomalies as well as concept drifts, the results suggest that our model can sufficiently detect anomaly from data stream and update model timely to fit the latest data property.. Index Terms: About four key words or phrases in alphabetical order, separated by commas 07 30 2021 233 241 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijrte.org true 10.35940/ijrte.B6294.0710221 https://www.ijrte.org/wp-content/uploads/papers/v10i2/B62940710221.pdf M.Tech degree, Department of Computer Science and Engineering, Faculty Of Engineering and Technology, Agra College, Agra (U.P.), India. Kajal Gupta R K Sharma Associate Professor, Faculty Of Engineering and Technology, Agra College, Agra (U.P.), India. In the world of today, computers have begun to rule the people as the machines carry out practically every work that people can accomplish. Scene classification is one such concept that becomes increasingly important when robots replicate the actions of a human being Scene categorization may be done on interior or exterior scenes using various extraction techniques, as well as categorization of indoor and outdoor scenes in these two categories is more difficult. The methodology for the indoor/outdoor classification scene has the drawback of inadequate accuracy. This research aims to enhance the accuracy by using the Convolution Neural Network Model in VGG-16. This paper proposes a new approach to VGG-16 to classify images into their classes. The algorithm results are tested using SUN397- indoor-outdoor dataset & the tentative data reveal that the methodology proposed is superior to the existing technology for the scene classification of indoor-outdoor (I/U). 07 30 2021 242 247 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijrte.org true 10.35940/ijrte.B6297.0710221 https://www.ijrte.org/wp-content/uploads/papers/v10i2/B62970710221.pdf