Modules
The MixedEmotions big data platform is a software tool conceived to enable efficient processing of diverse data sources in order to identify sentiments and emotions in context. The main properties of the platform are the following:
- Multimedia data ingestion and processing. The platform is capable of collecting data from multiple modalities including text, audio and video. Processing includes the measurement of sentiments and emotions across modalities as well as recognition and linking of topics and entities.
- Scalability. The platform has “big data” capabilities - it can be scaled to match the volume of input data, from a few thousand items on a single machine to billions of items processed across the cloud.
- Multilingual data processing. The platform is prepared to handle data in multiple languages. It has high quality models for Spanish, English, Italian and Czech as well as models for more than 15 European languages built with the aid of machine translation.
The issue tracker on our GitHub repository to can be used for opening discussions on potential improvements to code and documentation.
The MixedEmotions platform modules list.
Module | Language(s) | Description | Links | ||
---|---|---|---|---|---|
Audio Emotion Analysis | Language Independent (French trained) |
![]() Estimating arousal, valence, age, gender, big 5 personality traits from audio. |
Run a Task | ||
Video Emotion Analysis | Language Independent |
![]() Face detection, arousal and valence for detected faces, body stance estimation. |
-- | Run a Task | |
Text Emotion Analysis (Distant Learning) | EN |
![]() Distant learning models for emotion detection based on Twitter hash tags for Ekmans six emotions. |
Run a Task | ||
Text Emotion Analysis (Lexical) | Multi-lingual |
![]() Uses English text emotion lexical resources to produce values for basic emotion categories. |
Run a Task | ||
Text Sentiment Analysis (Deep Learning) | EN |
![]() Sentiment analysis models using LSTM neural networks trained on Stanford treebank movie review data. |
Run a Task | ||
Text Sentiment Analysis (Sentiment140) | EN |
![]() Sentiment140 is an open source sentiment detection model. |
Run a Task | ||
Text Sentiment Analysis (Senti-WordNet) | EN, ES, IT, FR |
![]() Uses "Senti-WordNet", a text sentiment lexical resource, to produce sentiment values. |
Run a Task | ||
Text Sentiment Analysis (Rule and Lexicon Based VADER) | EN, ES |
![]() VADER is an open source sentiment detection algorithm utilising gramatical rules and sentiment lexicons. |
Run a Task | ||
Text Sentiment Analysis (Multiple) | EN, CS |
![]() Java wrapper around several sentiment analysis tools, that was created for MixedEmotions project. |
-- | ||
Text Suggestion Mining | EN |
![]() Models to identify suggestions in text. |
Run a Task | ||
Topic Extraction | ES |
![]() Spanish topic extraction |
Run a Task | ||
Entity Extraction | ES |
![]() Spanish entity extraction. |
Run a Task | ||
Entity Extraction and Linking | EN |
![]() Entity recognition and linking with Wikipedia entities. |
Run a Task | ||
Knowledge Graph | Multiple | Provides insights into relations between recognised entities using semantic knowledge from DBpedia. | Usecases | ||
Machine Translation | Multiple |
![]() This module provides machine translation from multiple European languages to English and vice versa. |
-- | Run a Task | |
Social Network Analysis (SCANER) | -- |
![]() Social network sampling and analysis. Returns several metrics of social influence and context. |
External Link | ||
Platform Orchestrator | -- | The code of this orchestrator will let users have a starting point on how to interact with the platform modules. | N/A | N/A | |
Marathon Configurations | -- | In this repository you will find Marathon configuration files for the different MixedEmotions modules that you can find in MixedEmotions’ Dockerhub. | N/A | N/A | |
Data Visualisation | -- | Kibi is a friendly – kept in sync – Kibana fork which add support for joins across indexes and external sources, tabbed navigation interface and more | N/A | N/A | |
Emotion Recognition Wrapper (Python) | -- | A sentiment and emotion analysis server in Python | N/A | ||
Emotion Recognition Wrapper (Java) | -- | A sentiment and emotion analysis server in Java | N/A | N/A |