Demographic attributes prediction is fundamental and important in many applications in real world, such as: recommendation, personalized search and behavior targeting. Although a variety of subjects are involved with demographic attributes prediction, e.g. there are requirements to recognize and predict demography from psychology, but the traditional approach is dynamic modeling on specified field and distinctive datasets. However, dynamic modeling takes researchers a lot of time and energy, even if it is done, no one has an idea how good or how bad it is. To tackle the problems mentioned above, a framework is proposed in this chapter to predict using classifiers as core part, which consists of three main components: data processing, predicting using classifiers and prediction adjustments. The component of data processing performs to clean and format data. The first step is extracting relatively independent data from complicated original dataset. In the next step, the extracted data goes through different paths based on their types. And at the last step, all the data will be transformed into a demographic attributes matrix. To fulfill prediction, the demographic attributes matrix is taken as the input of classifiers, and the testing dataset comes from the same matrix as well. Classifiers in the experiments includes conventional state-of-the-art ones and Extreme Learning Machine, a new outstanding classifier. From the results of experiments based on two unique datasets, it is concluded ELM outperforms others. In the stage of prediction adjustments, two kinds of adjustments strategies are proposed corresponding to single target attributes and multiple target attributes separately, where single target attributes adjustments strategies include: adjusting the parameters of classifiers, adjusting the number of classes of target attributes and adjusting the public attributes. And multiple target attributes adjustment utilizes the outputs of first prediction as the inputs of second prediction to improve the accuracy of the first prediction. The framework proposed in this chapter consumes less time compared with traditional dynamic modeling methods, and there is no need to fully study the knowledge in various subjects for researchers using the framework because of the regular patterns. In addition, adjustment strategies have no restriction on the datasets; hence it will be useful universally. However, in some cases, dynamic modeling has the advantage of precision, resulting in better accuracy, but the results from the framework proposed in the chapter could provide as a comparison. In this work, a universal demographic attributes prediction framework is proposed to work on a variety of dataset with Extreme Learning Machine (ELM). The framework consists of three main components: First, processing raw data and extracting attribute features depending on different data types; Second, predicting desired attributes by classification; Third, improving the accuracy of classifiers through various adjustment strategies. Two experiments of different data types on real world prediction problems are conducted to demonstrate our framework can achieve better performance than other traditional state-of-the-art prediction methods with respect to accuracy. abstract environment.