A Spectral Approach to Item Response Theory
This work addresses a key estimation challenge in item response theory, with applications in education testing and recommendation systems, but it is incremental as it builds on the well-established Rasch model.
The paper tackles the problem of estimating item parameters in the Rasch model, a fundamental model in item response theory, by proposing a new algorithm based on a Markov chain on an item-item graph, achieving consistency and favorable optimality properties with competitive performance on synthetic and real-life datasets.
The Rasch model is one of the most fundamental models in \emph{item response theory} and has wide-ranging applications from education testing to recommendation systems. In a universe with $n$ users and $m$ items, the Rasch model assumes that the binary response $X_{li} \in \{0,1\}$ of a user $l$ with parameter $θ^*_l$ to an item $i$ with parameter $β^*_i$ (e.g., a user likes a movie, a student correctly solves a problem) is distributed as $\Pr(X_{li}=1) = 1/(1 + \exp{-(θ^*_l - β^*_i)})$. In this paper, we propose a \emph{new item estimation} algorithm for this celebrated model (i.e., to estimate $β^*$). The core of our algorithm is the computation of the stationary distribution of a Markov chain defined on an item-item graph. We complement our algorithmic contributions with finite-sample error guarantees, the first of their kind in the literature, showing that our algorithm is consistent and enjoys favorable optimality properties. We discuss practical modifications to accelerate and robustify the algorithm that practitioners can adopt. Experiments on synthetic and real-life datasets, ranging from small education testing datasets to large recommendation systems datasets show that our algorithm is scalable, accurate, and competitive with the most commonly used methods in the literature.