Zarqa Alyamama is an Arabic social media analytics product that enables our customers to retrieve and expanded set to content using Arabic language query expansion, store and index content, then analyze it using many statistical and machine learning (ML) tools. One of the unique advantages of this product is the highly accurate Arabic sentiment classification.
The following are some of Zarqa Alyamama’s product features:
1. Collecting and storing Twitter data:
The data collecting admin is responsible for collecting a large amount of data from Twitter. The collected data then are used as seeds for different statistics and processing activities. We used the standard Twitter APIs (streaming APIs and REST APIs) in collecting and storing activates. Data retrieval admin has been built as dynamic as possible so we can collect tweets based on specific keywords, hashtags, users or events.
The reporting part is a complete web tool which provides insight to different statistics and information in a form of visual reports which can be gathered from the collected tweets:
The Monitoring part is a complete web tool which enables the user to monitor specific actions, events, keywords, hashtags and users, by building his alert criteria according to his needs.
4. Hashtag life cycle:
The hashtag life cycle reports give the user the ability to monitor and track the history and statistics information of a specific hashtag.
The trending reports give the user the ability to monitor top trends not only for one day as Twitter provides, but also for a previous time interval by using the huge amount of data we maintain in our storage systems.
6. Sentiment Analysis:
In the sentiment analysis section, users can get the sentiment of the people with respect to a specific topic and classify it into three main groups (positive, negative and neutral). The classification process can be obtained from three main tabs which differ from each other by the technology used in the classification. Zarqa Al Yamama uses three main ways to classify tweets based on their sentiment as follows: