کنفرانس مهندسی برق ایران

صفحه اصلی / سی امین کنفرانس بین المللی مهندسی برق

Multinomial Emoji Prediction Using Deep Bidirectional Transformers and Topic Modeling

نویسندگان :

Zahra Ebrahimian¹ Ramin Toosi² Mohammad Ali Akhaee³

1- دانشگاه تهران 2- دانشگاه تهران 3- دانشگاه تهران

کلمات کلیدی :

emoji prediction،social network analysis،topic modeling،BERT

چکیده :

Abstract—As social networks grow more prominent, emojis become highly popular with a huge number of users. The primary function of emojis is to fill in emotional cues that might otherwise be missing from textual communication. These days, emojis are considered to be a large part of popular culture around the world. As a consequence, proper use of emojis in text messages will make you more friendly. Recently, emoji prediction becomes a challenging task, due to the large number of classes and the lack of suitable datasets. In this paper, we propose a multimodal emoji prediction model using the contextual information and the visual information. We used the EfficientNetB7 network to extract information from the images. EfficientNetB7 gives us the 10 most likely classes for each image. Also, we used Latent Dirichlet Allocation (LDA) as topic modeling to find hidden topics in the text. The topics extracted by LDA are then combined with the BERT network to improve the performance of this network. On our collected dataset, we achieved 46% accuracy for five emojis and 36% accuracy for 10 emojis.

لیست مقالات

لیست مقالات بایگانی شده

بکارگیری یادگیری عمیق در ارزیابی به هنگام پایداری ولتاژ کوتاه مدت با استفاده از داده های اندازه گیری فازوری

امیرحسین باباعلی - محمدتقی عاملی

A Comprehensive Analysis of a Digital Control Strategy for Photovoltaic-Based Single-Phase Grid-Tied Inverter Systems

Soheil Hasani Sangani - Mohamad Reza Moslemnejad - Mojtaba Saeedi - Alireza Jalalitalab - Reza Beiranvand

HyperSpectral Image Classification using a 3D Convolutional Mixer Block

Sara Dianat - Mehran Yazdi

On spatiotemporal-aware deep neural networks for real-time video fire detection: empowering image-based models with temporal and spatial features of video

Mahdi Shamisavi - Sahar Eslami - Amir Jahanshahi - Morteza Rajabzadeh

Machine Learning-based Fundamental Stock Prediction Using Companies’ Financial Reports

Hossein Rezaei - Kamran Abdi - Mohsen Hooshmand

Tuning of SMC Parameters for 3-DOF Spatial Parallel Robot Based on Whale Optimization Algorithm

Saeed Firuz Bahr Afzal - Amir Hossein Hassanabadi

بررسی خواص حسگری نانوکامپوزیت پلی آنیلین / اکسید وانادیوم 〖(V〗_2 O_5) نسبت به گاز اتانول در دمای اتاق

محمد حسین سعادتمند - محمد حسین شیخی

Low Complexity Single-Snapshot DOA Estimation Using Adaptive Filtering

Mojtaba Amiri - Mohammadreza Nargesi - Ali Olfat

Robust Object Detection Against Adversarial Perturbations with Gabor Filter

Mohammad Parsa Karimi - Abdollah Amirkhani - Shahriar B. Shokouhi

Achieving a Wide Range of Voltage Gain in Three-Phase LLC Resonant Converter Using Hybrid Control of Variable Frequency and Variable Magnetizing Inductor

Saeed Ramezani darvish - Salar Sadeghian - Adib Abrishamifar

بیشتر

ثمین همایش، سامانه مدیریت کنفرانس ها و جشنواره ها - نگارش 42.8.0