Lucene Search Jobs

4 were found based on your criteria {{|number:0 }} were found based on your criteria

show all
  • Hourly ({{ jobTypeController.getFacetCount("0")|number:0}})
  • Fixed Price ({{ jobTypeController.getFacetCount("1")|number:0}})
Hourly - Expert ($$$) - Est. Time: Less than 1 week, Less than 10 hrs/week - Posted
Project is at the idea stage and i'm discussing it with potential partner but to go forward i would need to have some numbers on possible expenses. Can you estimate how many hours it may take to develop solr or lucene search engine restful service that would allow to index files (MS word, excel, pdf & html) in specific directories and also would take search request and return results with synopsis plus it would return documents with highlighted hits.
Skills: Lucene Search Apache Solr Java EE REST
Fixed-Price - Intermediate ($$) - Est. Budget: $200 - Posted
The goal of this project is to implement a special document indexing and search strategy using Pylucene (or a different searching library if you can convince us it is better). The code will be simple and short. The hard part is understanding lucene and Pylucene at a deep enough level to index things correctly. This needs to support indexes of at least 1M documents, where each document is approximately a 20 page Word document. The indexing class should be ConceptIndexer, which will have these elements (more detail attached): def __init__(index_name) - load index, or create if not exists def define_fields(field_def_dict) - field_def_dict is a map from a string name to a tuple The tuple, for now, is just a type: number, date, keyword - There are two special fields - ‘text’, which is indexed as normal lucene document text - ‘concepts’, which is indexed as normal lucene document text def index_document(field_dict, text, concept_dict) - field_dict maps from field name to value - text is stored in a special field ‘text’, which is indexed as a normal lucene document - concept_dict is a mapping from a concept name to an integer score (if necessary with PyLucene, you can add a ‘close_index’ method) A second class is ConceptSearcher, which needs these functions: def __init__(index_name, concept_map) - Load index - concept_map is a large mapping from lower-case words to an array of concept ids def search_docs(search_query, offset=0, max_results=100) - search_query should accept the standard lucene query formats: - - Before executing the search in lucene, call _rewrite_concepts_in_query(search_query) - Execute the query with offset and max_results constraints - Return the full documents as an array of JSON objects. def _rewrite_concepts_in_query(search_query) - This function will convert all ‘c:word‘ and ‘concept:word’ elements into a new string using a very simple algorithm I can generate a small dataset for you to index and search.
Skills: Lucene Search
Fixed-Price - Expert ($$$) - Est. Budget: $2,000 - Posted
I need to perform several lucene queries on a column of text within a CSV. The output must include a binary output that indicates if the text within a cell contains or does not contain each lucene search string. Ideally, I would like to be able to replicate this process in the future for non-technical analysts. For this reason, I believe a connection between Solr and Rapidminer would be a user friendly option, but I am open to other methods for executing this. Thank you very much! An example of the tags is here: An example of the data is here (and attached):
Skills: Lucene Search Apache Solr Elasticsearch Rapid Miner
Fixed-Price - Expert ($$$) - Est. Budget: $1,000 - Posted
Hello All, We are building a application for matching professionals with job Postings and jobs with professionals. For this we need matching enigne which can do clustering, indexing, Searching, matching, and then providing recommendations based on match score. When job seeker logs into application he should get job recommendation based on his profile without entring any search query and when recruiter logs in he should get candidates recommendations based on jobs he posted. We need to do matching in real time as the user logs in we have to find all matching jobs to his profile.We use sqlite database and we intended to use Elasticsearch or Solr to build this recommendaton/search engine. Your solution should be highly scalable, robust ,accurate and should work for variety of job types and resumes like any other job portal like Careerbuilder,linkedin or monster. Information we are collecting is : (Stored in SQLite) Data is highly structured and you wont have to work much on data preparation. JOB POSTING 1. Skills required for job. - This is array of skills. ex. Java, php,html,python,sql 2. Total years of experience required for job 3. Job description - Text paragraph (Need parsing to find token to boost search) 4. Desired Qualification - Tex Paragraph (Need parsing to find token to boost search) 5. Education qualification required (Low importance) 6. Job function (Low importance) (There are other paramters too like job location,salary. Those can be used to boost search and matching score. CANDIDATE PROFILE 1. Overall Skills -This is array of skills. ex. Java, php,html,python,sql 2. Total years of experience of candidate 3. Past experiences Experience 1 - No of years in this position (months) Skills used array - ex. Java, php,html,python,sql Job responsibilities: Text paragraph about job responsibilities Experience 2 - No of years in this position (months) Skills used array - ex. Java, php,html,python,sql Job responsibilities: Text paragraph about job responsibilities Experience 3 - No of years in this position (months) Skills used array - ex. Java, php,html,python,sql Job responsibilities: Text paragraph about job responsibilities and so on.. 4. Education qualification of candidates (Low importance) 5. Job function (Low importance) Based on information collected above we want you to build algorithm to recommend jobs to candidates and candidates to jobs.We are using PHP for our application so code in PHP is preferred but not necessary. We proposes Elasticsearch/Solr for clustering, matching and providing recommendations. Elasticsearch/Solr will be deployed outside our ready application. We can keep data in Sqlite database, but data has to be ingested in Elasticsearch/Solr in realtime in order to Elasticsearh/Solr can build an inverted index, make clusterization and perform faster search. Ready application will deal with Elasticsearch/Solr by means of HTTP requests, fetching recommendations and showing them to a user. Elasticsearch/Solr Should provide content-based and behaviour based,like this recomendations. The final solution will consist of: - set of code modules, scripts and instructions for Solr deployment and configuration; - client library with examples of making search/recommendation requests - documentation, describing applied algorithms,methods - Generate random set of jobs/seekers - Setup Elasticsearch/Solr, create schema, ingest data - Attribute-based (matching by job title, skills, years of experience, location) - Hierarchical Classification (requires hierarchical classification of job/industries) - More-like-this (matching by job description, previous experience) - Concept-Based (recommending concepts (clusters)) - Collaborative filtering (recommending jobs "liked/applied" by similar users) In future we may have similar work. So consider this as long term opportunity. Don't bid if you don't have background in Elasticsearch/Solr/Search-Information retrieval/machine learning or you haven't done similar kind of work before...
Skills: Lucene Search Apache Solr Elasticsearch Machine learning