SemEval-2017-Task-5

This repository contains the source codes produced by RiTUAL-UH team for participating in SemEval-2017-Task-5: Fine-Grained Sentiment Analysis on Financial Microblogs and News.

• RiTUAL-UH ranked 2nd in subtask-2 (News Headlines) and 6th in subtask-1 (Financial Microblogs).
• Using alternative metrics that incorporates company informations, RiTUAL-UH ranked 1st in both of the subtasks.

For more Details

More Details of the task could be found in the Overview Paper.

More Details about the system could be found in the System Description Paper.

Bibtex to cite this paper

@InProceedings{kar-maharjan-solorio:2017:SemEval,
  author    = {Kar, Sudipta  and  Maharjan, Suraj  and  Solorio, Thamar},
  title     = {RiTUAL-UH at SemEval-2017 Task 5: Sentiment Analysis on Financial Data Using Neural Networks},
  booktitle = {Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017)},
  month     = {August},
  year      = {2017},
  address   = {Vancouver, Canada},
  publisher = {Association for Computational Linguistics},
  pages     = {877--882},
  abstract  = {In this paper, we present our systems for the “SemEval-2017 Task-5 on Fine-
	Grained Sentiment Analysis on Financial Microblogs and News”. In our system,
	we combined hand-engineered lexical, sentiment and metadata features, the
	representations learned from Convolutional Neural Networks (CNN) and
	Bidirectional Gated Recurrent Unit (Bi-GRU) with Attention model applied on
	top. With this architecture we obtained weighted cosine similarity scores of
	0.72 and 0.74 for subtask-1 and subtask-2, respectively. Using the official
	scoring system, our system ranked the second place for subtask-2 and eighth
	place for the subtask-1. It ranked first for both of the subtasks by the scores
	achieved by an alternate scoring system.},
  url       = {http://www.aclweb.org/anthology/S17-2150}
}

Directory Description

• source_project : Pycharm project containing source codes written in Python

-- [experiments] : codes of the models

-- [prepare_data] : codes for creating sequences and other preprocessing operations

-- [features] : lexical and embedding feature extraction functions

• submission : Predicted Sentiment scores for the test data

Contact

Sudipta Kar email: skar3 AT uh DOT edu