As more and more online search queries come from voice, automatic speech recognition becomes a key component to deliver relevant search results. Errors introduced by automatic speech recognition (ASR) lead to irrelevant search results returned to the user, thus causing user dissatisfaction. In this paper, we introduce an approach, "Mondegreen", to correct voice queries in text space without depending on audio signals, which may not always be available due to system constraints or privacy or bandwidth (for example, some ASR systems run on-device) considerations. We focus on voice queries transcribed via several proprietary commercial ASR systems. These queries come from users making internet, or online service search queries. We first present an analysis showing how different the language distribution coming from user voice queries is from that in traditional text corpora used to train off-the-shelf ASR systems. We then demonstrate that Mondegreen can achieve significant improvements in increased user interaction by correcting user voice queries in one of the largest search systems in Google. Finally, we see Mondegreen as complementing existing highly-optimized production ASR systems, which may not be frequently retrained and thus lag behind due to vocabulary drifts.

Many recent advances in neural information retrieval models, which predict top-K items given a query, learn directly from a large training set of (query, item) pairs. However, they are often insufficient when there are many previously unseen (query, item) combinations, often referred to as the cold start problem. Furthermore, the search system can be biased towards items that are frequently shown to a query previously, also known as the 'rich get richer' (a.k.a. feedback loop) problem. In light of these problems, we observed that most online content platforms have both a search and a recommender system that, while having heterogeneous input spaces, can be connected through their common output item space and a shared semantic representation. In this paper, we propose a new Zero-Shot Heterogeneous Transfer Learning framework that transfers learned knowledge from the recommender system component to improve the search component of a content platform. First, it learns representations of items and their natural-language features by predicting (item, item) correlation graphs derived from the recommender system as an auxiliary task. Then, the learned representations are transferred to solve the target search retrieval task, performing query-to-item prediction without having seen any (query, item) pairs in training. We conduct online and offline experiments on one of the world's largest search and recommender systems from Google, and present the results and lessons learned. We demonstrate that the proposed approach can achieve high performance on offline search retrieval tasks, and more importantly, achieved significant improvements on relevance and user interactions over the highly-optimized production system in online experiments.

Modern online content-sharing platforms host billions of items like music, videos, and products uploaded by various providers for users to discover items of their interests. To satisfy the information needs, the task of effective item retrieval (or item search ranking) given user search queries has become one of the most fundamental problems to online content-sharing platforms. Moreover, the same query can represent different search intents for different users, so personalization is also essential for providing more satisfactory search results. Different from other similar research tasks, such as ad-hoc retrieval and product retrieval with copious words and reviews, items in content-sharing platforms usually lack sufficient descriptive information and related meta-data as features. In this paper, we propose the end-to-end deep attentive model (EDAM) to deal with personalized item retrieval for online content-sharing platforms using only discrete personal item history and queries. Each discrete item in the personal item history of a user and its content provider are first mapped to embedding vectors as continuous representations. A query-aware attention mechanism is then applied to identify the relevant contexts in the user history and construct the overall personal representation for a given query. Finally, an extreme multi-class softmax classifier aggregates the representations of both query and personal item history to provide personalized search results. We conduct extensive experiments on a large-scale real-world dataset with hundreds of million users from a large video media platform at Google. The experimental results demonstrate that our proposed approach significantly outperforms several competitive baseline methods. It is also worth mentioning that this work utilizes a massive dataset from a real-world commercial content-sharing platform for personalized item retrieval to provide more insightful analysis from the industrial aspects.

In recent years, opinion mining attracted a great deal of research attention. However, limited work has been done on detecting opinion spam (or fake reviews). The problem is analogous to spam in Web search [1, 9 11]. However, review spam is harder to detect because it is very hard, if not impossible, to recognize fake reviews by manually reading them [2]. This paper deals with a restricted problem, i.e., identifying unusual review patterns which can represent suspicious behaviors of reviewers. We formulate the problem as finding unexpected rules. The technique is domain independent. Using the technique, we analyzed an Amazon.com review dataset and found many unexpected rules and rule groups which indicate spam activities.

This paper aims to detect users generating spam reviews or review spammers. We identify several characteristic behaviors of review spammers and model these behaviors so as to detect the spammers. In particular, we seek to model the following behaviors. First, spammers may target specific products or product groups in order to maximize their impact. Second, they tend to deviate from the other reviewers in their ratings of products. We propose scoring methods to measure the degree of spam for each reviewer and apply them on an Amazon review dataset. We then select a subset of highly suspicious reviewers for further scrutiny by our user evaluators with the help of a web based spammer evaluation software specially developed for user evaluation experiments. Our results show that our proposed ranking and supervised methods are effective in discovering spammers and outperform other baseline method based on helpfulness votes alone. We finally show that the detected spammers have more significant impact on ratings compared with the unhelpful reviewers.

This paper studies structured data extraction from Web pages. One of the effective methods is tree matching, which can detect template patterns from web pages used for extraction. However, one major limitation of existing tree matching algorithms is their inability to deal with embedded lists with repeated patterns. In the Web context, lists are everywhere, e.g., lists of products, jobs and publications. Due to the fact that lists in trees may have different lengths, the match score of the trees can be very low although they follow exactly the same template pattern. To make the matter worse, a list can have nested lists in it at any level. To solve this problem, existing research uses various heuristics to detect candidate lists first and then applies tree matching to generate data extraction patterns. This paper proposes a generalized tree matching algorithm by extending an existing tree matching algorithm with the ability to handle nested lists through a novel grammar generation algorithm. To the best of our knowledge, this is the first tree matching algorithm that is able to consider lists. In addition, it is well-known that there are two problem formulations for Web data extraction: (1) pattern generation based on multiple pages following the same template, and (2) pattern generation based on a single page containing lists of data instances following the same templates (each list may use a different template). These two problems are currently solved using different algorithms. The proposed (single) algorithm is able to solve both problems effectively. Extensive experiments show that the new algorithm outperforms the state-of-the-art existing systems for both problems considerably.

Evaluative texts on the Web have become a valuable source of opinions on products, services, events, individuals, etc. Recently, many researchers have studied such opinion sources as product reviews, forum posts, and blogs. However, existing research has been focused on classification and summarization of opinions using natural language processing and data mining techniques. An important issue that has been neglected so far is opinion spam or trustworthiness of online opinions. In this paper, we study this issue in the context of product reviews, which are opinion rich and are widely used by consumers and product manufacturers. In the past two years, several startup companies also appeared which aggregate opinions from product reviews. It is thus high time to study spam in reviews. To the best of our knowledge, there is still no published study on this topic, although Web spam and email spam have been investigated extensively. We will see that opinion spam is quite different from Web spam and email spam, and thus requires different detection techniques. Based on the analysis of 5.8 million reviews and 2.14 million reviewers from amazon.com, we show that opinion spam in reviews is widespread. This paper analyzes such spam activities and presents some novel techniques to detect them

Mining of opinions from product reviews, forum posts and blogs is an important research topic with many applications. However, existing research has been focused on extraction, classification and summarization of opinions from these sources. An important issue that has not been studied so far is the opinion spam or the trustworthiness of online opinions. In this paper, we study this issue in the context of product reviews. To our knowledge, there is still no published study on this topic, although Web page spam and email spam have been investigated extensively. We will see that review spam is quite different from Web page spam and email spam, and thus requires different detection techniques. Based on the analysis of 5.8 million reviews and 2.14 million reviewers from amazon.com, we show that review spam is widespread. In this paper, we first present a categorization of spam reviews and then propose several techniques to detect them.

It is now a common practice for e-commerce Web sites to enable their customers to write reviews of products that they have purchased. Such reviews provide valuable sources of information on these products. They are used by potential customers to find opinions of existing users before deciding to purchase a product. They are also used by product manufacturers to identify problems of their products and to find competitive intelligence information about their competitors. Unfortunately, this importance of reviews also gives good incentive for spam, which contains false positive or malicious negative opinions. In this paper, we make an attempt to study review spam and spam detection. To the best of our knowledge, there is still no reported study on this problem.

This paper studies the problem of identifying comparative sentences in text documents. The problem is related to but quite different from sentiment/opinion sentence identification or classification. Sentiment classification studies the problem of classifying a document or a sentence based on the subjective opinion of the author. An important application area of sentiment/opinion identification is business intelligence as a product manufacturer always wants to know consumers' opinions on its products. Comparisons on the other hand can be subjective or objective. Furthermore, a comparison is not concerned with an object in isolation. Instead, it compares the object with others. An example opinion sentence is "the sound quality of CD player X is poor". An example comparative sentence is "the sound quality of CD player X is not as good as that of CD player Y". Clearly, these two sentences give different information. Their language constructs are quite different too. Identifying comparative sentences is also useful in practice because direct comparisons are perhaps one of the most convincing ways of evaluation, which may even be more important than opinions on each individual object. This paper proposes to study the comparative sentence identification problem. It first categorizes comparative sentences into different types, and then presents a novel integrated pattern discovery and supervised learning approach to identifying comparative sentences from text documents. Experiment results using three types of documents, news articles, consumer reviews of products, and Internet forum postings, show a precision of 79% and recall of 81%. More detailed results are given in the paper.