Our I.T. consultation firm (Information Age Consulting) represented by its social media analytics arm (Social Intelligence) implemented a Sentiment Analysis System for Modern Standard Arabic and Kuwaiti dialect. The system is built using Finite State Transducers which is a technique widely used in the field of Computer Science and Computational Linguistics for Language processing. The system is available as an API (Application Programming Interface) which utilizes Repustate Sentiment Analysis Engine (repustate.com). This API can be used as a cloud-based web service and can be integrated with any social media analytics platform or application to provide sentiment analysis functionality for Modern Standard Arabic and Kuwaiti Dialect. The API produces its output in JSON format. Our Kuwaiti dialect lexicon contains 450000 words with their sentiment polarity (Positive, Negative, and Neutral). Our Modern  Standard Arabic Lexicon contains around 8 million words in Modern Standard Arabic with their sentiment polarity.  Our electronic lexicon of modern standard Arabic is driven from modern electronic corpora and it contains words currently used by Arab speakers and writers in books, newspapers, and other media. This lexicon contains many words that are not available in the Traditional Standard Arabic dictionary. These include words related to politics and science in addition to newly coined terms. The lexicon contains the possible morphological forms (inflections and derivations) for the Modern Standard Arabic words (Examples: كتب، ويكتبان، كتبوا، سيكتبن، كاتب، المكتوب).  Our Modern Standard Arabic and Kuwait dialect lexicons also encode the Morpho-Syntactic features of words using the ISO Standard Lexical Markab Language. Our lexicon of Modern Standard Arabic is one of the largest lexicons in the industry for modern Standard Arabic in terms of size, coverage, and updated word forms. Besides, there is no sentiment lexicon available in the industry for the Kuwaiti Dialect, which make our Kuwaiti dialect lexicon the first of its kind terms of size and coverage.  Our Kuwaiti dialect lexicon contains the possible morphological forms (inflections and derivations) of Kuwaiti Dialect words (Examples: ريال ، ريالين، رياييل). These words are the Kuwaiti dialect counterparts of the standard Arabic words :   رجل ، رجلان، رجال respectively. Kuwaiti dialect also uses some standard Arabic words that are identical to words available Modern Standard Arabic. These standard words are available in our Modern Standard Arabic lexicon. Our Kuwaiti dialect lexicon undergoes continuous updates. We are updating it with more words driven from social media channels using corpus processing and Natural Language Processing applications. Besides, our sentiment analysis API recognizes emoticons and emoji symbols and give them high weight in the overall sentiment score.

We have also developed a social media analysis application “Social Intelligence Analyzer” which allows governmental and private sector entities to analyze dialogues and activities on social media through analyzing the linguistic content of what is being said about a specific word, phrase, or brand. This system works as an add-on to Pulsar Social Media Monitoring Systemand  Brandwatch Analytics System in order to analyze the Arabic posts retrieved by Pulsar Social Media Monitoring System and Brandwatch Analytics system using the techniques of Text Mining, Data Analysis, Computational Linguistics, and Natural Language Processing. The system is optimized to understand and analyze posts written in Standard Arabic Language and Kuwaiti Dialect and it can detect the important insights and indicators from posts after eliminating noisy data through recognizing and distinguishing between noisy Arabic words and relevant/important words. To apply this, our system uses an advanced Morphological Analysis system for Arabic language. In this respect, the system detects the Top Words, Top Bigrams (two successive words), Top Trigrams (three successive words), Top Quadgrams (four successive words). The system also provides sentiment analysis for the analyzed posts through integrating with our Arabic Sentiment Analysis API. This system is currently used by the Kuwaiti Parliament to analyze Twitter dialogues and acquire insights and indicators about public opinion and public trends regarding the parliament issues and performance. More information about our social media analysis system can be found on this page:

http://intelligence.social/siteENG/index.php/advanced-analytics

 

You can have more information about our social media analytics services by visiting the web site of our social media analytics arm (Social Intelligence):

http://intelligence.social/siteENG

We are currently working on an Arabic Taxonomy and Sentiment Analysis project in collaboration with SAS (sas.com). The project aims at building the Arabic language processing capabilities that will allow SAS Social Media Analytics System to use Modern Standard Arabic and Kuwaiti dialect for Sentiment Analysis by developing Standard Arabic and Kuwaiti dialect taxonomies. The taxonomies are built using Finite State Transducers a widely used technique in the field of Computer Science and Computational Linguistics for Language processing.