Cyber-security has evolved to a key priority for Qatar and all nations over the world. Malicious actors from anywhere misuse the cyberspace to perpetrate various crimes such as phishing, Cyber-blackmailing, Cyber-bullying, and communicating or planning terrorist attacks using social media. For instance, there is a tendency from these cybercriminals to use similar writing styles in their messages, which makes it possible for security experts to detect and stop these threats using natural language processing techniques (NLP) as explained in the IBM Security Intelligence online news source. In that context, this project aims at developing author profiling resources and tools for the Arabic language and applying them to cyber-security as an instrument for fighting against cyber-crimes. More specifically, author profiling can be useful for forensics investigations to narrow the set of potential authors when receiving a threat message. While a few research efforts on author profiling have recently started in Europe and the USA, there is extremely little research that targets the Arabic language (only the two works found: Abbasi and Chen (2005) and Estival et al (2008)). Our project will bridge that gap and pave the way for further research on Author profiling for Arabic. Our research will address different characteristics for author profiling such as the age, the gender, the native language, the language variety, and the author interests. Furthermore, we will work on detecting deception and irony in Arabic text, which is necessary for distinguishing serious content from humoristic content. The collected data can be used to intercept suspect events, like the preparation of a terrorist act in a social media stream. The proposal has three main components: the language resources, the author profiling tools, and the application scenarios. For the resources, we plan to create a suite of seven large-scale annotated Arabic author profiling resources. This includes annotated corpora for: (1) author interests, (2) gender detection(e.g. males vs. females), (3) age detection (e.g. youth vs. adults), (4) native language (e.g. native vs. non-native), (5) Arabic dialect variety (e.g. Qatari Arabic vs. Iraqi Arabic), (6) Irony detection (e.g. sarcasm vs. sincerity) and (7) deception detection in online message (e.g. truth vs. lies). Based on the collected resources we will build a set of tools using a combination of linguistic features and machine learning techniques in order to determine automatically several aspects of a given text written by an anonymous author. We will build tools to automatically infer the interest topics of the author (e.g. sport, religion, extremism, violence), to detect the gender and the approximate age of the writer, whether he is a native Arabic speaker or not and also the Arabic dialectal variety used. Furthermore, we plan to profile the author for his degree of sincerity or irony (e.g. if the text is likely to be based on truth or lies). Following a service-oriented architecture, these tools will be provided as modular and reusable services that can be easily used in third parties applications. These tools will be integrated to provide comprehensive Arabic author profile reports. The third project components are the application scenarios. We will take cyber-security as our main scenario. In that context, the author profiling tools will support both government organizations and internet/telecom providers in tracking down cyber-crimes. Beyond cyber-security, the author profiling tools can also be used by businesses to improve their marketing and service processes. In the context of social customer relationship management, author profiling can be used by Arab companies for marketing intelligence (e.g., to better understand customer needs and feedback) and also for market segmentation (e.g., to better target each customer group). Finally, author profiling can be used in the medical domain to identify individuals with dangerous psychological profiles and possibly detect and prevent suicide attempts.