Answer yes/no queries in search engines
نویسنده
چکیده
According to the query logs in search engines, yes/no questions are issued frequently. One example yes/no question is “is Obama the 42nd president of the United States”, and the expected answer is “No”. However, the most advanced search engines only return a list of ranked documents containing the keywords within the query. Then users have to go through those returned documents in order to figure out one simple answer “No”. People may argue that this query contains semantic and it can hardly be handled by search engines. Nevertheless, current search engines have the ability to process most of the natural language questions. For instance, issuing the query “who is the 42nd president of the United States” on Google, the answer is “Bill Clinton”. Therefore, current search engines should have the ability to handle most of the yes/no queries. Based on the answer of “who is the 42nd president of the United States”, the answer of “is Obama the 42nd president of the United States” should be easily obtained. Thus, current search engines can answer most of the yes/no questions. The ability of the answering yes/no questions will enhance the power of current search engines and will significantly improve the users’ search experiences. Yes/no queries can be divided into two categories objective yes/no queries (OYN questions) and subjective yes/no queries (SYN questions). • The answers of OYN queries are simple “Yes” or “No” with evidences, which can be obtained based on existing facts. For example the answer to the OYN question “is universal studios hollywood open now?” is “Yes” if the following two facts are gained. (1) The open time of Universal Studios Hollywood is 9AMĺC10PM from Monday to Friday, and (2) it is 9:30AM Monday now. The answer becomes “No” if now is “11PM Monday”. • The answers of SYN queries are (Y es,Py) and (No,Pn) pairs with evidences, where Py and Pn are the probability of the answers to be “yes” and “no” respectively. There are no facts for such queries, but user subjective opinions. For example the answer to the SYN question “is ios better than android” is “(Yes, 55%), (No, 45%)” based on analyzing users’ comments, tweets and blogs. Architecture of processing yes/no queries. Figure 1 shows the architecture of processing yes/no queries. The Query Analyzer analyzes the query to identify the type. One possible solution is to check whether it contains subjective terms like comparison terms, e.g., “better than” and “good”. But it is not an accurate measurement, as such terms can sometimes be a non. E.g., query “is good an English word?” is a OYN query even it contains term “good”. To improve the accuracy of query type identification, query logs should be utilized. Based on query logs, if a similar query “is bad an English word” is an OYN query, then “is good an English word?” is an OYN with high probability. The Context Manager collects users’ time-spatial information and also relevant documents from search engines via keyword search. Time-spatial information is served as filter condition, while keyword search returns only relevant context of queries. For example, in order to answer the query “is it rainy today”, the following three information should be obtained: (1) today’s date, e.g., 8/1/2016, which is the time information, (2) user’s location, e.g., New York, which is the spatial information, and (3) the weather of New York on 8/1/2016. The OYN/SYN Processor processes the OYN and SYN queries based on the corresponding context and query types, and returns answers in corresponding format. Query Analyzer SYN OYN Processor SYN Processor Search Engines Spatial/time Monitor Context Manager OYN OYN context SYN context Yes/No + evidences (Yes,Py), (No, Pn) + evidences query is Obama the 42nd president of the United States No Obama is the 44th president of the United States Clinton is the 42nd president of the United States More about Bill Clinton and Barack Obama Figure 1: Architecture of processing yes/no queries. Results exhibit Figure 2(a) shows an example output format for a OYN query “is it rainy today”, while Figure 2(b) demonstrates an example result of an SYN query “is ios better than android”. The outputs directly answer yes/no queries. Users can also see more details to see the evidences. Query Analyzer SYN OYN Processor SYN Processor Search Engines Spatial/time Monitor Context Manager OYN OYN context SYN context Yes/No + evidences (Yes,Py), (No, Pn) + evidences query is Obama the 42nd president of the United States No Obama is the 44th president of the United States Clinton is the 42nd president of the United States More about Bill Clinton and Barack Obama
منابع مشابه
مدل جدیدی برای جستجوی عبارت بر اساس کمینه جابهجایی وزندار
Finding high-quality web pages is one of the most important tasks of search engines. The relevance between the documents found and the query searched depends on the user observation and increases the complexity of ranking algorithms. The other issue is that users often explore just the first 10 to 20 results while millions of pages related to a query may exist. So search engines have to use sui...
متن کاملGoogle Scholar’s Ranking Algorithm: An Introductory Overview
Google Scholar is one of the major academic search engines but its ranking algorithm for academic articles is unknown. We performed the first steps to reverse-engineering Google Scholar’s ranking algorithm and present the results in this research-in-progress paper. The results are: Citation counts is the highest weighed factor in Google Scholar’s ranking algorithm. Therefore, highly cited artic...
متن کاملSemantic search engine: A survey
The semantic search engines have some advantages on the web search engines from the users view. In this fast life everybody need the answer for their queries very fast .In this scenario semantic searching engines will be helpful. It deals with the actual meaning of the queries. The tremendous growth in the volume of data or the information lead the traditional search engines to get the answers ...
متن کاملBelief dynamics in web search
People frequently answer consequential questions, such as those with a medical focus, using Internet search engines. Their primary goal in performing these searches is to revise or establish beliefs associated with one or more outcomes. Search engines are not designed to furnish answers, and instead provide a ranked list of documents, some of which contain answers. People's prior beliefs about ...
متن کاملPrefix search with a lie
The problem of performing a search procedure when some of the answers to queries may be erroneous has been recently investigated by many authors (see [3,4, 2, 11). One of the main questions in this domain was stated by S. M. Ulam [S]: what is the minimal number of yes-no queries sufficient to find an unknown object in an n-element set if at most one answer may be erroneous? This problem was fir...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2017