# Anserini Installation & Usage ### Producer: hhchen --- ## Installation: 1. Install Java 11 (JDK/OpenJDK) For those who install Java the first time or already install another version, please set your PATH & JAVA_HOME 2. Install Maven (>= 3.5) 3. clone the anserini from [GitHub](https://github.com/castorini/anserini) 4. go to the directory and compile it. --- ## Usage: Regrettably, though it is a fabulous tool, it does not have a complete document about itself. (Maybe because of its frequent update.) Try to follow the examples it provides, and you may find these links useful: 1. [Anserini Docs](https://github.com/castorini/anserini/tree/master/docs) 2. [Common Indexing Options](https://github.com/castorini/anserini/blob/master/docs/common-indexing-options.md?fbclid=IwAR1TFz42AENZfb9pFgKigYYHU2faaAVfBU1yPjX_GUufaqYp1jUFQEsctjc) --- # Dive in to the Case: To show my kindness, let’s check this example together: [MS MACRO with BM25](https://github.com/castorini/anserini/blob/master/docs/experiments-msmarco-passage.md?fbclid=IwAR0uO9R5Z8qyyiJ9YbeBUkl7utL3Kp6-0N5-mhxwxnzNgkBoquUkWwhFyCg) --- # Pyserini: In most cases, anserini is a kind of command-line tools. However, feel free to use it in a Python program: [Pyserini](https://github.com/castorini/pyserini) --- # Murmur as an End: We are going to have a historical hard Project 3: [RecSys Challenge 2020](http://www.recsyschallenge.com/2020/), which is an industrial-grade IR problem. You may need such a tool to help you address with such ****ing large dataset, or you may find that you spend a whole week waiting for preprocessing.