This is not entirely "spam". It's aggregation of content. I checked the text on some of these sites and it's all scraped from authority sources. I know some of these sites link back to the original, while some don't. They use no more than 2-3 paragraphs from one source, but they tend to make big pages by showing at least 10-15 sources on one page.
First, there's keyword research and grouping. Without it, it's hard to believe you can make something decent. Then you need an API in the backend, which is fed by one or more scraping APIs (if you don't want to do the scraping by yourself, which means handling proxies, search engine changes in responses etc).
In the front, you can even have Wordpress and construct the pages with the JSON response from the API. You can cache the response and so the WP will become even faster. But most of these sites are using a custom solution for frontend.
So it's not spam. Maybe there's an aggressive interlinking to help indexing, but that's hardly an issue when you have wikipedia with all that interlinking. If you think about it, each google result page can be categorized as spam, since all the content is from external sources.
In the end, that's no rocket science. The hardest part is to automate keyword research and grouping. As soon as I'm done with my current project, I'll give it a try.